Kenny, Paul --- "Australia's Capital Gains Tax Discount: More Certain, Equitable And Durable?" [2005] JlATaxTA 11; (2005) 1(2) Journal of The Australasian Tax Teachers Association 38

	Home \| Databases \| WorldLII \| Search \| Feedback Journal of the Australasian Tax Teachers Association

The Ralph Committee, consisting of three leading businessmen, John Ralph (Chairman), Rick Allert and Bob Joss, was established in 1998 to make recommendations on reforms to the Australian tax system. The Committee sought to achieve three national taxation objectives: optimising economic growth, promoting equity and promoting simplicity and certainty.^[1] Additionally, the Committee considered that fiscal adequacy was important as the reforms were designed to achieve revenue neutrality.^[2]

In 1999 the Ralph Committee recommended sweeping reforms to the Australian income tax system. Its final report consisted of eight parts^[3] and made 280 recommendations. Many of these have since passed into law in a staggered series of stages since 1999. Relevantly, numerous Capital Gains Tax-related (CGT) recommendations were made and many of these have also passed into law. In accordance with the Ralph Committee’s four tax policy criteria set out above (equity, economic efficiency, simplicity and fiscal adequacy), this paper evaluates the CGT discount^[4] for individuals, trusts, superannuation funds and other related reforms^[5] that flowed from the 1999 Ralph Report. The paper finds that the CGT discount has greatly undermined equity and fiscal adequacy. Further, this policy appears to have created significant economic distortions and contributed to complexity.

II THE TAX POLICY CRITERIA

Fiscal adequacy refers to the ability of the taxation laws to finance government expenditure. Fiscal adequacy is a fundamental requirement for a tax system given the government’s need for revenue to ensure good governance.^[6]

CGT plays a vital role in fiscal adequacy since it produces a small but significant amount of tax revenue. CGT produced revenue of $5330 million in 1900–2000,^[7] $4412 million in 2000–01^[8] and $3657 in 2001–02.^[9] Additionally, CGT reinforces the tax base and prevents tax avoidance since taxpayers are prevented from converting ordinary income into tax-free capital gains.

The Ralph Report asserted that equity is important: ‘equity is a basic criterion for community acceptance of the tax system’.^[10] Tax equity is generally defined in terms of horizontal equity and vertical equity.^[11] Horizontal equity demands equal treatment for people in similar circumstances.^[12] This requires the determination of a tax base to measure ‘similar circumstances’ so that an appropriate amount of tax be imposed on a taxpayer. Accordingly, most commentators^[13] have defined the tax base in terms of a taxpayer’s ‘ability to pay’.

Thus horizontal equity requires those having an equal ability to pay to bear equal burdens of tax.^[14] As horizontal equity concerns the equal treatment of equals, as a corollary, vertical equity is required to ensure that tax imposed on people in different circumstances is also fair. Most countries have progressive rates of income tax^[15] so as to try to ensure that a person with a greater ability to pay, pays not only more tax, but at a higher income tax rate. Vertical equity requires both progressive income tax rates and a tax based on the ‘ability to pay’.

The Ralph Report^[16] noted that economic efficiency is vital since the tax system ‘can significantly influence the efficiency with which Australia’s natural resources, capital and labour are used’. Ideally the tax system should have a neutral impact on the economy so as to maximize economic efficiency. There is no shortage of literature on the economic virtues of such a neutral tax system and the ideal of a comprehensive income tax base^[17]. Cooper’s^[18] literature survey found that economic income has long remained a popular tax base ideal with western countries. Indeed since the 1980s there has been a long-term international trend towards comprehensive income taxation.^[19] Further, the Ralph Report’s business taxation design principles advocated that ‘the tax base adopted should be as close as possible to comprehensive income’.^[20] Similarly, the Asprey Report^[21] and Draft White Paper^[22] preferred this tax base. The results from large scale surveys of economists further illustrate the wide support for this tax base. Behrens^[23] surveyed 450 registrants of a 1971 National Taxation Conference and found that tax professionals overwhelmingly chose income tax as the fairest tax. Slemrod^[24] reported on a 1994 survey of economists that similarly found a strong preference for income and wealth taxation.

In regards to simplicity, the Ralph Report noted that ‘[a] major consideration in the formulation of the Review’s recommendations has been to remove anomalies and inequities between the treatment of economically similar transactions. This will allow significant simplification of the tax system’.^[25]

Simplicity, though, is an attribute that appears to have never sat very well with taxation laws,^[26] and is a difficult concept to define.^[27] There is, however, general agreement that simplicity is considered in terms of the compliance costs of taxpayers and the administration costs of government.^[28] Simplicity can, theoretically at least, be measured by estimating these costs, known as operating costs, and dividing this amount over the amount of tax revenue. Simplicity thus improves where the operating costs or this ratio falls.

Compliance costs can be defined as the costs ‘incurred by taxpayers, or third parties such as businesses, in meeting the requirements laid upon them in complying with a given structure and level of tax’.^[29] These costs will include the costs of keeping records, preparing taxation financial statements and taxation returns, obtaining tax advice, undergoing tax audits, tax planning and disputes. Taxes, though, can provide a number of benefits to taxpayers that may offset these costs.

Managerial benefits are provided as a result of improved business decision making flowing from tax law compliance.^[30] That is, the record keeping and financial information requirements of the tax laws provide taxpayers with better information to make business and investment decisions. Additionally, taxpayers obtain a benefit since these compliance costs are generally tax-deductible^[31] under s 8-1 and s 25-5.

There will also be cash flow benefits^[32] to taxpayers who are able to defer the payment of tax. Compliance costs can thus be represented by the following equation:

Taxation administration can be categorised into four types of government activities: tax policy, design and planning; tax law drafting and enactment; Australian Taxation Office; and tax dispute resolution.^[33] Administration costs can thus be represented by the following equation:

These policy criteria, however, are both complex and highly controversial.^[34] Firstly, there exists no known method of quantifying each of the criteria so as to facilitate a comparison. Secondly, there is no universal method for weighting each of the criteria. Given these limits, it is considered that all of these policy objectives are important.

Ideally the ‘optimal CGT’ structure maximises the equity, efficiency, simplicity and fiscal adequacy criteria. Whilst it is widely accepted that comprehensive income is the ideal income tax base^[35] commentators find that compromises must be made to this ideal in designing tax laws^[36] given the conflicting tax policy objectives. As Stiglitz noted,^[37] an optimal tax is ‘the one that maximizes social welfare, in which the choice between equity and efficiency best reflects society’s attitudes toward these competing goals’.

III THE CGT DISCOUNT AND RELATED RALPH REPORT CGT REFORMS

The Ralph Report recommendations 18.2 and 18.3^[38] resulted in the new Div 115 of the Income Tax Assessment Act 1997 (ITAA 1997) that replaced the former averaging and indexation concessions (noted below). Division 115 provides a CGT discount to individuals, trusts and complying superannuation entities. However, the CGT discount does not apply to companies. Prior to this reform, net capital gains were generally taxed at a taxpayer’s marginal income tax rate. The Ralph Report provided the following rationale for this reform:^[39]

The Review’s recommendations for capital gains taxation are designed to enliven and invigorate the Australian equities markets, to stimulate greater participation by individuals, and to achieve a better allocation of the nation’s capital resources. In the first three or four years of the new regime there is likely to be considerable extra turnover on Australian equity markets as equity holders respond to reduced lock-in by realigning their portfolios. Even in the medium to longer-term, the Review expects a heightened level of realisations activity amongst individual shareholders and CIVs [Collective Investment Vehicles]. The Review’s recommendations in regard to scrip-for-scrip rollover will also stimulate significant turnover in the wake of expected increased takeover activity (see Recommendation 19.3).

A number of stepped rate scales to provide relief that increased with the holding period of the asset were discussed in A Platform for Consultation (page 291). Such an approach has the appeal that it provides a reward for ‘patient capital’. However, there is an inherent tension between rewarding patient investors and seeking to free up capital markets. Realisation-based capital gains tax systems generally suffer from a tendency to lock asset holders into less than optimal positions. Providing any further reward for delaying realisation (for example, by means of a stepped rate related to holding period) would, in some cases, exacerbate the lock-in effect.

In consultation with business — especially the venture capital industries — it has also become clear to the Review that start-up ventures proceed through various stages, often needing to have capital restructurings even after only a few years in order to fund further development. A stepped rate scale would not appropriately reward shorter term investment at each such stage of the venture’s development.

For these reasons, the Review has decided to recommend a broad form of CGT relief which has only a limited relationship to the period of holding of the asset.

The Review’s recommendation of an exclusion of 50 per cent of capital gains for eligible assets held for a year or more by individuals will increase significantly the attractiveness of investing in capital-gains-bearing assets by individuals (see Recommendation 18.2). It will reduce the effective top marginal rate on capital gains income to 24.25 per cent.

Section 115-5 states that a discount capital gain is a capital gain that meets the following basic requirements:

1. The capital gain must be made by an individual; or a complying superannuation entity; or a trust.

2. The capital gain must result from a CGT event happening after or on 21 September 1999.^[40]

3. The capital gain must have been worked out using a cost base that excludes indexation.^[41]

4. The capital gain must result from a CGT event happening to a CGT asset that was acquired by the entity making the capital gain at least 12 months before the CGT event.^[42] The following ten CGT events though do not obtain the CGT discount: D1, D2, D3, E9, F1, F2, F5, H2, J2 and J3.^[43]

Also, there are two anti avoidance rules for the 12-month holding period. Section 115-40 provides that where the CGT event occurred under an agreement you made within 12 months of acquiring the CGT asset the CGT discount will not apply. Section 115-45 excludes a CGT discount for capital gains from certain disposals of equity interests in companies or trusts where the taxpayer could not have otherwise accessed the CGT discount if the taxpayer directly owned the underlying assets of the company or trust.

The discount percentage for an amount of a discount capital gain is 50 per cent if the gain is made by an individual or by a trust and 33.33 per cent if the gain is made by a complying superannuation entity.^[44]

Special rules for trusts apply so that the CGT discount is appropriately traced through to beneficiaries receiving capital gains and so that company beneficiaries do not receive the CGT discount.^[45]

It is also noted that access to the CGT discount was expanded to include assets realised by listed investment companies on or after 1 July 2001.^[46]

In conjunction with the introduction of the CGT discount a number of related CGT reforms were implemented. These reforms are set out below.

Ralph Report recommendation 18.1(a)^[47] resulted in the removal of the CGT averaging concession. Previously, under CGT averaging, individuals and certain trustees could average their net capital gains to reduce the bunching effect of an accumulated capital gain^[48]. The net capital gain for the income year was generally divided by five to provide the marginal rate of tax that would apply to the capital gain when added to other income. That rate of tax (for the one fifth of the capital gain) was multiplied by five to calculate the total gain. Ralph Report asserted:^[49]

[i]n order to deliver such a deep reduction in effective rates within the constraint of revenue neutrality, the Review has necessarily had to identify other features of the current regime which do not contribute to the objectives of encouraging investment or removing inflexibilities in the capital markets.

Australia’s averaging provisions were identified early as contributing little to these aims while reducing revenue substantially. They are used by a section of the asset-holding community to reduce capital gains taxation to zero, or near to zero, while others who are not in a position to engineer the same benefit carry the burden of taxation at close to their full marginal rate. This results in considerable inequity. In practice, investors facing high marginal rates of tax remain locked in to a significant extent rather than realise and, where realising, have to secure a much higher return on the reduced capital available for reinvestment. Some individuals, of course, achieve a very low tax rate on capital gains through the averaging provisions and that may encourage more investment by such people. But the Review is not convinced that overall efficiency is best promoted by present provisions.

Were averaging to remain for a period after announcement, the Review expects that many taxpayers would seek to structure their affairs in order to take full advantage of averaging in 1999–2000 while it remained available. Accordingly, the Review recommends removal of averaging in relation to any gain realised on sales contracted from the date of announcement (Recommendation 18.1(a)). This may create some complexity in the transition year for certain taxpayers and in the administration of the tax law. But the potential disruption to collections otherwise justifies this one-time complexity.

In accordance with this recommendation^[50] the CGT averaging system was abolished, effective from 21 September 1999.

The Ralph Report recommendation 18.1(b)^[51] resulted in the freezing of CGT indexation at 30 September 1999. Prior to this reform, the CGT provisions freely permitted a taxpayer who owned a CGT asset for at least 12 months to index the cost base of the asset for inflation in calculating a capital gain.^[52] The indexation commenced in the quarter when the expenditure on the cost base was incurred, and ceased in the quarter when the CGT event occurred.^[53]

As a result of this recommendation, these rules were amended to deny indexation of the cost base for CGT assets acquired from 21 September 1999.^[54] Additionally, the indexation amount of the cost base of CGT assets acquired at or before 21 September 1999, and disposed of after that time, is frozen as at 21 September 1999.^[55] The Ralph Report explained:^[56]

The freezing of indexation adjustments after the September quarter 1999 adjustment is the other major adjustment recommended by the Review to achieve revenue neutrality (Recommendation 18.1(b)).

The case for freezing indexation is not as clear-cut as for averaging. Though indexation provides a significant reduction in effective rate for many taxpayers, this is probably not well recognised, especially amongst foreign investors. Indeed, the perception has been that the Australian tax system imposes tax at full income tax rates. Such misperceptions are not easily corrected and a change in the form of concession to something more akin to the types of concession available abroad would, in the Review’s judgment, be more effective in attracting investors to Australian assets.

Notably, indexation still applies for pre-21 September 1999 assets but cannot be used in addition to the CGT discount.^[57]

As noted above, this paper evaluates the CGT discount and related reforms in regard to the tax policy criteria of fiscal adequacy, equity, economic efficiency and simplicity.

IV FISCAL ADEQUACY

The following table sets out net capital gains and CGT revenue before and after the September 1999 Ralph CGT reforms.^[58]

Australian net capital gains and CGT revenue 1999–2002

The levels of capital gains and CGT collections have fallen in the years following the Ralph CGT discount in 1999–2000. This may, however, be also due to other factors such as the decline in the stock market. It is clear though that the CGT discount has proven to be very costly to tax revenue as the following Treasury estimates illustrate:^[59]

Estimated tax expenditures from the CGT discount 2001 to 2006

The CGT discount has an enormous influence over CGT collections as seen by the size of the revenue loss of $1.8 billion in 2001–02 compared to CGT revenue of about $3.6 million in 2001–02.

In contrast, the reforms that removed CGT averaging^[60] and froze indexation^[61] have been revenue positive although the quantum of revenue raised is unknown.

V EQUITY

The CGT discount seriously undermines both horizontal and vertical equity as demonstrated by the following examples.

X who earns interest income of $50 000 and pays $11 322 income tax (including Medicare levy) in the income tax year ended 30 June 2004 will be in a horizontally equitable position with Y who realises a capital gain of

$50 000 and pays $11 322 income tax. However, if only 50 per cent of Y’s capital gain is taxed, horizontal equity will be breached as Y only pays $3897 income tax and X pays $11,322 income tax, yet both have the same ability to pay.

Also, the CGT discount breaches vertical equity as demonstrated by the following examples.

A high income earner, M, at the top marginal income tax bracket of 48.5 per cent (including Medicare levy), earns $70 000 of capital gains and pays tax of $20 757 income tax in the income year ended 30 June 2004. A taxpayer, N, with a top marginal income tax bracket of 31 per cent (including Medicare levy), earning $50 000 of salary pays $11 322 income tax. Thus vertical equity is maintained. However, if M is only taxed on 50 per cent of the capital gain, M will pay income tax of $7197 whilst N pays $11 322. Thus vertical equity is breached.

Further, if both taxpayers receive a discounted capital gain, the high income earner obtains a greater benefit from this exemption as seen below.

M, a high income earner at the top marginal income tax rate of 48.5 per cent, receives a $10 000 50 per cent discounted capital gain and thus obtains a tax saving of $2425 as a result of the CGT discount. N, on a 31.5 per cent marginal income tax rate, receives a $10 000 50 per cent discounted capital gain and thus only saves $1575 of income tax.

As the above examples show, the CGT discount has an ‘upside down effect’ in providing the greatest benefit to asset holders. Given the high concentration of wealth in Australia^[62] this preference results in significant losses to vertical equity. The benefit of the CGT discount is highly skewed since 10 per cent of the population have 43 per cent of the total wealth and 50 per cent have 90 per cent of the wealth.^[63] The bottom half of the population only have 10 per cent of the wealth, whilst the poorest 10 per cent have no wealth at all.^[64]

However, the reforms that removed CGT averaging and froze CGT indexation, improved horizontal equity and vertical equity since these concessions were only available to a limited number of CGT asset holding taxpayers.

Tax preferences damage tax integrity as they provide taxpayers and tax advisers with both the rationale and opportunity for tax avoidance^[65] and tax evasion.^[66] Additionally, it may lead to the corruption of tax administration.^[67]

A number of researchers^[68] have found that the belief that tax laws are unfair encourages the acceptability of avoidance and evasion. Pederick provides a telling analysis of the impact of tax preferences on community attitudes:^[69]

If the tax system and its administration are seen as to press heavily on some but lightly on other taxpayers in essentially the same circumstances, with essentially the same tax paying capability, then the confidence of the citizenry in the fairness and justness of the system, of their government, erodes. Cynicism grows apace and a race not to be left out of the tax minimisation derby, by hook or by crook, infects the body politic. Since government in our societies truly is “of the people, by the people and for the people”, to escape collection of one’s share of the costs of our joint enterprises is to rifle the common treasury, to take from one’s neighbour. A society that breeds an attitude that one’s duty is wholly to one’s self and not to the community, not to one’s country, when it comes to taxes, is a society with a cancer at work. It is, moreover, a cancer that threatens the quality of life on a wider front and, indeed, perhaps the long term stability of the society it infects.

This suggests that Australia’s CGT exemptions^[70] such as the CGT discount would have a profound impact on community attitudes. Thus, fully taxing capital gains would appear to greatly stem such cancer in community attitudes, although other tax exemptions would also need to be closed to effectively change attitudes and behaviour.

Tanzi found evasion is likely to occur where evasion opportunities exist.^[71] Other researchers have noted that tax advisers take full advantage of any ambiguity in tax laws. Porcano et al^[72] found that tax returns prepared with tax assistance, especially prepared by Certified Practicing Accountants (CPA) and lawyers had much higher levels of non-compliance than returns that were self-prepared. This could be due to interpretational problems—as Ayres et al^[73] found, advisers adopt an aggressive stance in ambiguous tax deduction cases. Also, Grbich^[74] noted that tax planners use time-honoured legalistic games to exploit the language of the law. Tax officers are mesmerised by the legalistic arguments and occasionally by superficial economic rhetoric.^[75]

Certainly CGT exemptions such as the CGT discount provide opportunities to transform income into exempt or concessionally taxed capital gains. Also, they provide tax advisers with the opportunity to exploit the ambiguities in the income–capital distinction.

Indeed, the Draft White Paper^[76] and the Australian Senate Finance and Public Administration References Committee Inquiry^[77] into the Business Taxation Reform both observed that exempting capital gains opens one of the greatest tax loopholes. A view shared by Burman.^[78] Australia’s income tax experience well illustrates such tax minimisation and evasion, as follows.

There is much evidence in Australia’s pre-CGT era of taxpayers and tax advisors exploiting the CGT exemption by transforming income into tax-free capital gains. For example, in the 1920s property development companies were established with the profits retained in the company which was subsequently placed in liquidation.^[79] In the absence of a CGT or any specific taxing provisions,^[80] the profits were then paid out as tax-free distributions of capital. The Treasurer said:^[81]

This has been done almost entirely for the purpose of enabling the shareholders to escape income tax on the liquidation dividends, because the courts have held that liquidation dividends are distributions of assets, and, in the present form of the income tax law, they cannot be classed as income even to the extent to which they have been distributed out of profits.

In the 1970s many companies engaged in a dividend-stripping process of tax avoidance.^[82] Companies with profits where no income tax was paid were acquired by promoters and then subsequently stripped of assets leaving the company unable to pay any taxes due. The vendor shareholders of the company received a tax-free capital payment for their shares whilst the promoters went on to strip assets. This practice was aided by the absence of a CGT that would assess the capital gains.^[83] The Explanatory Memorandum to the Taxation (Unpaid Company Tax) Assessment Act 1982 Part I explains that:

[a] typical scheme of pre-tax company profit stripping would ordinarily involve a sale of all (in isolated cases practically all) of the shares in a company (the ‘target company’) which had successfully traded for a substantial part of the income year and which had, up until the implementation of the scheme, current year profits on which a contingent company tax liability existed. In addition, if the target company concerned was a private company for income tax purposes it would in due course become liable to pay undistributed profits tax in the event that it failed to pay a dividend of a specified proportion of its profits within 10 months after the end of the income year.

The trading activities of the target company would first have been transferred to another entity (company or trust) controlled by the former owners of the company and the target company's assets reduced to cash or other liquid form. It would be a condition of the scheme promoter that all liabilities of the company except its actual or contingent tax liabilities be paid or indemnified by the vendor-shareholders.

The former owners of the company would be paid a price for their shares that was fixed on the basis of the value of the company's assets, not taking into account the contingent tax liability on company profits. This capital sum would however have been reduced to reflect the fee charged by the promoter or other stripper.

By further processes the target company would be stripped of its liquid assets (e.g., by the making of a loan that could not be repaid) and thus rendered incapable of meeting the company tax liability in due course assessed to it. It is this unpaid company tax that is the subject of the Bill.

There were yet other situations in which the ownership of companies was sold and the companies rendered incapable of paying their income tax.

Grbich estimated that revenue losses through avoidance and evasion from the bottom of the harbour schemes in the 1970s amounted to a minimum of $10 000 million in total.^[84]

In the 1980s the provision of lease incentives by owners of rental properties in the central business districts of Australian cities appeared to be another way in which taxpayers sought to exploit the absence of CGT. The Australian Taxation Office (ATO) described such arrangements in taxation ruling IT 2631:^[85]

The incentives take many forms, including large upfront cash payments, non-cash items such as top of the line motor vehicles or boats, expensive paintings, holiday packages, rent-free or rent-discounted periods for the leased premises or for premises in other cities, free fit-outs of the premises, payment of removal costs or for the surrender of the existing lease, interest-free loans, or a combination of these incentives.

The introduction of CGT into Australia in 1985 greatly improved the integrity of the tax laws. As well as taxing capital gains it has helped maintain levels of other tax revenue since ordinary income is prevented from being converted to tax-free capital gains. However, given the numerous exemptions,^[86] capital gains are only lightly taxed. The introduction of the CGT discount^[87] in 1999 further diluted the effectiveness of CGT as discussed below.

Under the CGT discount with no limit on negative gearing, the interest paid on investment loans is fully deductible. Thus a dollar of interest expense that produces $1 of capital gain reduces the investor’s taxable income by 50 cents. Deferral of capital gains under a realisation CGT^[88] offers an even greater profit, since the interest is fully deductible in the current period when it is incurred, yet the capital gains are deferred until realisation.

The CGT discount was expected to have a huge impact of tax arbitrage. The government’s estimate^[89] of $100 million per annum tax revenue losses was considered to be too low. Evidence from the Senate inquiry^[90] pointed to a massive loss of tax revenue. At the time of the Ralph Review, Krever found that the reforms would have a revenue leakage of hundreds of millions of dollars.^[91] Evans considered the reforms could transform the positive revenue estimate of $350 million into a significant revenue leakage of $5.5 billion in a worst-case scenario.^[92] Early Treasury estimates have clearly proven to be too low as the tax revenue lost on the CGT discount for individuals and trusts amounted to about $1.8 billion.^[93]

Additionally, CGT exemptions can lead to the corruption of tax administration. The Petroulias affair demonstrates how a tax loophole can result in such corruption. Petroulias, a former Assistant Tax Commissioner with the Australian Taxation Office, is subject to legal action^[94] over allegedly improperly receiving money for providing favourable private binding rulings to tax planners in respect of superannuation deductions. As Sandford noted, such behaviour is likely to severely damage taxpayer compliance.^[95]

It is apparent that such a CGT preference such as the CGT discount has a significant and adverse impact on community attitudes and encourages tax minimisation. Taxing capital gains at the taxpayers marginal income tax rate would be a vast improvement on the current regime.

The current Australian housing price bubble appears to be partially fuelled by preferential CGT treatment as investors sought to take advantage of the personal residence exemption, CGT discount and negative gearing.^[96] Relevantly, Sandford noted that preferential CGT treatment for housing creates inflated prices.^[97] This is unfair to purchasers, and especially for first home buyers who do not benefit from a price appreciation associated with a CGT exemption.

Since the 1999 Ralph Report, Australia has witnessed a surge in real estate investment as seen by the following Productivity Commission estimates of median city housing prices over the period 1998–2003.^[98]

This table illustrates the price surge over the period 1998–2003 with a 66 per cent average increase in city housing prices. This equates to an increase of 13.2 per cent per annum.

Commentators such as Warren^[99] have noted the investment advantages of appreciating investment property, given the benefit of a 50 per cent CGT discount on any capital gain and the immediate deductions for expenses including interest. Similarly, Bassanese^[100] noted that the benefit of the CGT discount for individual investors (as well as of other tax benefits such as negative gearing and capital allowances) led to a ‘stampede into investment properties in recent years’. He provided the following example from Reserve Bank of Australia’s (RBA) submission to the Productivity Commission inquiry into first home ownership:^[101]

A high income earner who partly draws against existing home equity to obtain 100 per cent interest only, financing for a $400,000 newly constructed investment property.

Even with a rental yield of only 3.5 per cent versus a loan interest rate of 6.5 per cent, the RBA shows that the investor need only outlay $81 in after tax dollars per week to maintain the investment after allowing for various other tax benefits like negative gearing and depreciation.

What the RBA does not say, but can be inferred, is that the implied annual capital gain required to break even on such an investment is less than 2 per cent, or long run expected inflation. It’s still true if we assume a higher long run interest rate of 7.25 per cent.

Also pertinent to this discussion is the Commission’s public inquiry into the affordability and availability of housing for first home buyers.^[102] Relevantly, its discussion paper found that CGT was an influence over housing prices:^[103]

To the extent that currently low housing affordability reflects cyclical price pressures, this will eventually be reversed. (Evidence of market cooling is already emerging.) However, there is a role for policy to address any forces that cause prices to be excessive over time.

On the demand side, interactions within the taxation system between negative gearing, capital gains provisions and marginal income tax rates lend impetus to investment demand when prices are rising.

Additionally, Bassanese^[104] observed that the preferential tax treatment for housing in Australia had caused great difficulty for new entrants to the housing market given the influx of investors buying housing. He observed,^[105] ‘Australia now has twice as many landlords among its taxpayers as North America, and six times as many as the United Kingdom, on a per capita basis’. Further, the lack of housing affordability appears to be a real problem given the high levels of homeless people^[106] and high levels of people renting.^[107]

There is no known distributional modelling for the impact of the CGT discount in Australia. However, the distributional consequences can be inferred from estimates of the concentration of Australian wealth noted above. Therefore, it appears that the CGT discount appears to be one of the contributing factors to Australia’s growing inequality of income and wealth.^[108] This is supported by a study by Saunders^[109] who noted that ‘taxes and transfers may have exacerbated the existing pressures on inequality, particularly in recent years’.

A growing divide between low income and high income households leads to unequal access to jobs, education, community services and justice.^[110] Further, this results in an increase in disharmony, social conflict and crime.^[111] As Stillwell found, this trend resulted in the following network of interlocking socio economic inequalities.^[112]

Additionally, a study by Bloomquist^[113] found a statistically significant correlation between income inequality and tax evasion consistent with behavioural and economic theories of tax evasion. Thus Bloomquist concluded that a tax policy that widens income equality would also be likely to lead to greater levels of tax avoidance and evasion.^[114]

It is a principle of international tax that there should not be any discrimination between capital gains accruing to residents and non-residents.^[115] Failing to tax capital gains though would contravene this principle and mean that Australia would not get its fair share of international CGT since most countries in the Organisation for Economic Cooperation and Development (OECD) tax capital gains.^[116] Such an exemption would mean that equity is breached between domestic investors investing in Australia and overseas. Domestic taxpayers investing in Australia will enjoy tax-free capital gains yet be taxed on foreign capital gains. Equity is also breached between foreign investors who invest in Australia because they would pay no CGT in Australia yet be subject to CGT in their country of residence.

With the great divergence in the way OECD countries tax capital gains,^[117] it is clear that the current Australian CGT rules would not always achieve international tax equity. However, the CGT discount may be justified on international equity grounds if other countries similarly discounted capital gains.

Notably, in 1999 (and prior to the CGT discount) Evans and Sandford made a comparison of CGT rates for selected countries including Australia.^[118] They found that Australia had the highest CGT rate of 48.5 per cent (including Medicare levy) compared to the maximum rate in the United Kingdom of 40 per cent, Canada and United States of America with maximum CGT rates of about 30 per cent, Ireland’s 20 per cent flat tax rate and New Zealand’s zero CGT rate.^[119]

A number of arguments are put forward for the preferential taxation of capital gains on equity grounds. Some of the concerns are based on the definition of ‘equity’ and the appropriateness of a comprehensive income tax base. Other arguments are based on the practical problems inherent in the operation of a CGT.

The traditional view of horizontal and vertical equity is debated by commentators^[120] who argue over the meaning of equity. The concept of vertical equity poses difficulties. How does a society agree on the level of taxes for taxpayers who have different abilities to pay? These difficulties are seen in the divergence of personal income tax rates between countries.^[121] Although most countries appear to embrace vertical equity by adopting progressive income tax rate structures, the ideal degree of vertical equity is highly contentious.^[122] The goal of horizontal equity, however, demands the taxation of capital gains and, as asserted previously,^[123] this goal should not be negotiable.

CGT causes bunching of a capital gain that has accrued over a period of time. Thus it is asserted that it is unfair to tax these gains in the one income year as this pushes a taxpayer into a higher tax bracket.^[124] Bunching would appear to impact on many taxpayers given the number of taxpayers who receive capital gains and who have taxable incomes of less than $62 500 in 2003–04 year.^[125] Below $62 500 of taxable income taxpayers face progressive marginal income tax rates. At $62 500, the top marginal income tax rates cuts in and thus bunching is not a problem for taxpayers with taxable incomes over this level in the 2003–04 year.

As discussed previously, Australia’s CGT originally provided averaging of capital gains for individuals to alleviate the problem of bunching,^[126] but was removed for capital gains derived from 21 September 1999 with the introduction of the 1999 Ralph CGT reforms.^[127]

The bunching argument flounders on the fact that capital gains constitute income in the same way as salary and wages form part of a person’s ability to pay and should be taxed. In any event many taxpayers will not be greatly affected by bunching because most CGT is paid by taxpayers in the top marginal tax bracket.^[128] A taxpayer in the top marginal tax bracket on a 48.5 per cent income tax rate will pay the same CGT liability regardless of whether the gains are incurred annually or whether a long-term capital gain is lumped into one tax year. Additionally, bunching will not have a great impact since the average net capital gain for individuals with taxable incomes below $50 000 is only $3073 in the 1999–2000 year.^[129] Additionally, under Australia’s CGT the bunching effect is offset by the benefit of CGT deferral until realisation.^[130]

It is argued^[131] that the capital gain on an asset includes an inflation component and that CGT should not apply since this component of the gain is not a real gain. As noted previously, Australia’s CGT initially featured a system of inflation indexation.^[132] However, following the introduction of the 1999 Ralph CGT reforms, indexation was restricted to assets acquired at or before 21 September 1999, with the indexation factor being frozen as at 30 September 1999.

There appear to be strong reasons for full indexation. If the tax system is not adjusted for inflation, reforms may not achieve their distributional goals. That is, the taxation and social security systems may not adequately achieve a redistribution of income from the wealthy to the poor.

For example, failing to index the income tax brackets for inflation will cause low and middle income taxpayers to be pushed into higher tax brackets as their incomes rise, even where there is no change in their real incomes. The 1975 Inflation and Taxation Report^[133] recommended that personal tax indexation, via consumer price index be applied to the taxable income brackets and deduction limits. It found that low income large families were the most adversely effected by inflation.^[134]

Further, Jorgenson and Kun-Young modelled the effects of the United States Tax Reform Act 1986 that broadened the income tax base and reduced tax rates, finding that much of the potential gain to welfare dissipated through the failure to index the income tax base for inflation.^[135] They argued that the tax system should be indexed for inflation otherwise welfare gains from tax reform will fail to materialise.^[136] Additionally, Musgrave^[137] asserted that an equitable concept of income must be in real terms.

However, inflation actually impacts more heavily on other types of capital income such as interest, rent and dividends over the course of a tax year than capital gains.^[138] Given that these other types of income are taxed without any inflation adjustment, then the inflation component of capital gains should also be taxed since it would be highly inequitable to differentiate between taxpayers receiving different types of income.

It is also argued^[139] that CGT is unfair since it taxes the same asset twice thus capital gains should be exempted or concessionally taxed. As the value of an asset is equal to the discounted value of the expected future cash flow generated by the asset, then any capital appreciation must flow from an increase in the expected future cash flow.^[140] This results in double taxation as the increased cash flow will be taxed when it is actually realised.^[141]

This argument misses the point that there are two parts to economic income from a business or investment asset. The first is the income flow and the second is the increase in asset value. Both contribute to a taxpayer’s ability to pay and should form part of the tax base, as seen in the following examples.

A person X discovers a gold mine which has a 20-year life. If, instead of mining the gold, that person sells the mine to Y and pays CGT, in this situation there will be no double taxation. Person Y will amortise the cost of the purchase against mining income and/or the cost base and this will offset any capital gains on the subsequent sale of the mine by Y. There is no double taxation.

Alternatively, if the asset is a share in a public company with an indefinite life, the result is the same. Person V sells the share to W, and instead of receiving dividends, V pays CGT. There will not be any double taxation for W since W includes the cost of the share purchase against sales proceeds of the share thus offsetting the capital gain.

Krever points out that the double tax argument is based on the following unrealistic assumptions:^[142]

(4) that changes in the value of an asset are never due to changes in the discount rate;

Overall, this argument does not appear to be valid in Australia since the purchase price of an asset would generally be included in the asset’s cost base.^[143]

Since the Courts have generally maintained that capital gains are not ordinary income, being unexpected, non-recurring and unlike gains that flow from productive effort, it has been suggested that they should not be taxed or should be concessionally taxed.^[144] It is submitted that there is no basis though for formulating taxation policy on the basis of taxpayer expectations rather than on a taxpayer’s ability to pay. If unexpected gains are exempt this will only lead to the artificial creation of many such unexpected gains by taxpayers. Further, this argument simply does not apply to modern investment practices which well anticipate appreciation in the value of assets such as marketable securities and real estate. Alternatively, it could be argued that unexpected capital gains should be taxed more heavily, since the gains are unexpected, the tax will not be missed.

It is also argued that introducing or modifying CGT is unfair to those people who have purchased assets on the understanding that the current CGT provisions would not change.^[145] However, other taxpayers who do not obtain such tax preferences end up paying higher taxes in order to make up the lost tax revenue. This is obviously more unfair. Thus this argument does not carry much weight.

To minimise the risk to tax revenue from the manipulation of capital losses by strategic trading, a CGT must limit the deductibility of capital losses to capital gains, rather than other income.^[146] Thus the Australian CGT regime restricts capital losses —such losses can only be offset against capital gains.^[147] It is argued that this is unfair as capital gains are included in assessable income but there is only limited tax relief for capital losses.^[148] On this basis capital gains should be exempted or concessionally taxed.

Indeed this restriction means that undiversified investors may pay substantial tax if their investment prospers yet they may obtain no benefit for failures.^[149] This unfairness though is offset to some extent by the benefit of deferral of tax on capital gains provided by a realisation CGT.^[150] The CGT discount, however, reduces the value of capital losses where such losses have to be offset against capital gains on assets that qualify for the CGT discount.^[151] In this situation an individual taxpayer having capital losses will only obtain benefit of 50 per cent of the capital losses being offset against capital gains. Thus the CGT discount reduces the value of deductible losses.

It is clear that the current CGT discount breaches both horizontal and vertical equity. The available evidence suggests that this exemption greatly undermines tax integrity and the redistribution goal of income taxation. The arguments for the CG|T exemptions lack merit. However, the reforms to remove CGT averaging and to freeze indexation are welcomed as these have provided some assistance to horizontal and vertical equity.

VI ECONOMIC EFFICIENCY

There is widespread support in Australia for tax policies that have a neutral impact on economic behaviour. All of the recent Australian income tax inquiries, the Asprey Report,^[152] the Draft White Paper^[153] and the Ralph Report,^[154] agreed, at least conceptually, on the neutrality advantages of a comprehensive income tax base. Relevantly, under this tax base all forms of income should be taxed similarly. In particular, since capital gains are a form of capital income, they should be taxed in similar manner to other types of capital income such as interest, dividends and rent, and this would allow capital to achieve the highest possible rate of return. To exempt or lightly tax capital gains will lead to a misallocation of capital.^[155]

The following analysis of capital gains and other types of capital income seeks to demonstrate the economic impact of tax preferences such as the CGT discount in Australia’s CGT regime. Reported capital income from Australian taxpayers for the 1999–2000 year is set out in the following table.

Dividends provide the greatest source of savings returns followed closely by net capital gains and gross interest. Notably, rent provides little reported net return, a mere $220 million.^[156]

A complex picture emerges on the taxation of Australian capital income given the myriad of tax rules, income tax rates^[157] and CGT exemptions that apply to the various types of taxpayers: individuals, trusts, companies and superannuation funds. Within the limits of this paper a full analysis of these differences is not possible. Rather, the paper focuses on the enacted Ralph Report CGT discount.

Interest, dividends and rent are included in a taxpayer’s assessable income^[158] and taxed annually at the taxpayers’ marginal rate of income tax. However, a CGT discount^[159] applies to the CGT rate for individuals, trusts and superannuation funds. The following examples illustrate the tax advantages for capital gains.

At the top marginal income tax rate of 48.5 per cent, $100 of annual interest income from bonds will generate income tax of $48.50. A capital gain on an asset held by an individual for a year or more will be taxed at the discounted CGT rate of 50 per cent. This equates to income tax of only $24.25 (50 per cent of $48.50). Under this simple analysis, and ignoring other factors, taxpayers would prefer capital gains from income, given the lower tax rates. Thus other assets such as interest bearing deposits would need to produce higher rates of return to attract capital.

The above analysis though ignores inflation, CGT deferral and differences in asset characteristics that also impact on the asset holding decision.^[160] As is evident from the example in Table 5 below, a capital appreciating asset is subject to an even lower effective tax rate^[161] than a bond in the presence of inflation since the tax is deferred until realisation. Indeed the higher the inflation rate the greater the difference between the effective tax rates for bonds and appreciating assets.^[162] Inflation affects capital gains less than interest, dividends and rent.^[163] Thus, as noted previously,^[164] there exists a strong case for indexing income and expenses.^[165]

The following table compares two investments held by an individual taxpayer at the top marginal income tax rate of 48.5 per cent. The taxpayer has $100 of bonds yielding 10 per cent per annum over five years and $100 of capital-appreciating shares paying no dividends but yielding 10 per cent per annum constant growth over five years. All of the net yield from the bonds and shares are reinvested and both investments are realised after five years.

With the after tax yield reinvested, the bonds will be worth $128 and shares $146 at the end of five years. The shares will have a capital gain of $61 and CGT of $14.80.^[166] Thus the after tax value of the shares is 13.8 per cent more than the bonds’ value. This difference of $17.70 arises from the benefit of the CGT discount ($14.80) and the greater compounding of asset growth due to the deferral of CGT ($2.90).

Introducing a 4 per cent inflation rate into the above example, the bonds and shares both provide a real pre tax return of 5.77 per cent.^[167] The real after tax return on the bond though is 1.6 per cent^[168] and for the shares it is 3.75 per cent.^[169] The effective tax rate is the percentage of reduction in the real annual return from an asset caused by taxes.^[170] For bonds the effective tax rate is 100 per cent^[171] and for the shares it is 62 per cent.^[172] This reflects the higher tax rate on bonds and the benefit of deferral on shares. Moreover, the longer the holding period the larger the benefit of deferral.

Overall, taxing capital gains at normal tax rates minimises the discrepancies between the taxation of the types of capital income. A realisation CGT though still distorts decisions given the advantage of deferral and the impact of inflation. However, an annual accruals CGT has other problems.^[173] Further, indexation should only be applied to capital gains if it is also extended to the entire income tax system. Introducing a CGT discount further exacerbates the differences. Financial markets and taxpayers are very adept in exploring such preferences as seen by Australia’s CGT experience, discussed previously.^[174]

(b) Australian Evidence of the Misallocation of Resources from CGT Preferences

Prior to September 1985, Australia generally excluded capital gains from income tax. Numerous commentators noted the significant economic costs of such a policy. Head^[175] outlined the distortions caused by Australia’s selective tax concessions for retirement savings, imputed rent and capital gains. He found that far from promoting growth, these provisions have only served to produce an extremely conservative pattern of saving and investment unsuited to the requirements of a dynamic and growing economy.^[176] Groenewegen^[177] similarly noted the economic distortions stemming from the failure to tax capital gains. Dixon^[178] also concluded that the gross misdirection of resources into property with low yields and tax avoidance schemes in the late 1970s would not have occurred if a CGT had been in place in Australia.

Additionally, Freebairn^[179] asserted that failing to tax capital gains penalises investment in risky ventures and favours large conglomerates over small enterprises. Krever^[180] noted that the seventy years experience in Australia of exempting CGT suggested that CGT exemptions are an inefficient tool for directing investment towards entrepreneurial activities. Rather it leads to low risk investment in real estate or blue chip shares.^[181] Further, Krever noted that a CGT preference extends beyond productive Australian assets to foreign shares, real estate and Australian holiday homes.^[182] These arguments would similarly be relevant to a partial exemption such as the CGT discount.

Following the commencement of CGT in 1985, the introduction of the CGT discount in 1999 imposed certain economic distortions. As noted previously,^[183] this policy has contributed to an over investment in housing. Warren asserted that the CGT discount and other property tax concessions have resulted in the Australian economy’s over reliance on the property sector for economic growth (given the strong multiplier effect of the building boom) and have led to a property price bubble.^[184] This creates further economic problems since a policy change that removes the capitalised value of tax concessions would in turn damage the building industry, lead to loan defaults and lower consumer confidence.^[185] Also, the RBA’s Deputy Governor, Glenn Stevens^[186] warned that the housing boom was a threat to the durability of Australia’s long economic upswing.^[187]

(c) International Evidence of the Misallocation of Resources from CGT Preferences

Sandford^[188] noted that the 1960s and 1970s saw burgeoning government expenditures and increasing government intervention in many countries, based on over-optimistic assumptions of how the government could influence society and the economy for the better. Concerns over the economic and social costs from the tax distortions flowing from these interventionist policies heralded a worldwide change in economic philosophy in the 1980s.^[189] This philosophy saw a revival of belief in the efficacy of markets and the need for tax neutrality.^[190] Countries pushed back the role of the state’s influence in the economy and deregulated, freed exchange rates, privatised, promoted competition and improved the efficiency of the public sector.^[191] Sandord found that this not only greatly affected the capitalist world but also ‘undermined the socialist planned economies’.^[192] This movement illustrates the widespread support for neutral tax policies and suggests the acceptance of the goals of comprehensive income taxation, and the full taxation of capital gains.

A number of United States analysts have noted economic distortions created by CGT preferences. McGee^[193] concluded that a CGT rate cut is nearly always more beneficial to the existing firm than the new start-ups. In many cases the cut will decrease new firm investment.^[194] McIntyre et al^[195] found that a CGT preference is incoherent and will remain that way until Congress defines the scope of the preference and matches it to a public policy.

McIntyre et al queried why a CGT preference should encourage a hoarder of gold coins or a passive real estate investor.^[196] Gordon and Slemrod’s^[197] analysis of 1983 income tax returns suggested that moving to uniform tax rates on real capital income would raise revenue and improve efficiency.

Kay and King^[198] reviewed the impact of the United Kingdom income tax laws finding that they have a great impact on household portfolios. Over the last 25 years the proportion of personal wealth held in the form of privileged assets such as houses, life insurance policies and pension funds, rose from 29 per cent to 64 per cent.^[199] Over the same period the personal holdings of shares and marketable securities fell by almost 75 per cent.^[200] Thus they concluded that preferential tax treatment for gains made from certain assets has had a major impact on the allocation of resources.^[201]

Sandford^[202] found that tax concessions to house owners raised the value of homes, providing capital gains to owners, but reduced the capacity of would-be purchasers. Also, concessions given to housing, life insurance and pensions in the United Kingdom divert savings to safe investments rather than riskier and innovative investments.^[203] He asserted that this provides the strongest argument for neutrality, via a progressive consumption tax or a comprehensive income tax.^[204]

The South Africa Department of Finance^[205] noted the many distortions created in the economy from the failure to tax capital gains as one of the primary reasons for the introduction of CGT in 2001.

In 1991 the Swedish government undertook massive policy changes in its ‘tax reform of the century’.^[206] The government reduced the income tax rates resulting in a revenue loss to the government of 6 per cent of gross domestic product.^[207] In return for these massive tax cuts the government broadened the tax base with a new system of taxing capital income and by broadening the value added tax and eliminating other loopholes.^[208] Auerbach reviewed the economic literature on the impact of the Swedish tax reforms and found that these reforms reduced tax planning, induced a shift away from owner-occupied housing and improved economic efficiency.^[209] He concluded that these reforms had a positive albeit modest impact. He asserted:^[210]

Agell, Englund and Sodersten conclude that Sweden which started with a very complex tax structure and high marginal tax rates, improved economic efficiency through its reform aimed at simplification and base broadening. But they also found that the changes in real behaviour, such as saving and labor supply, were modest, and the transaction costs considerable. Thus, the reform was beneficial, particularly with Sweden’s initial tax system, but not the panacea some might have predicted. Engen and Skinner conclude that tax policy can impact at least short run economic growth by enough to make an important difference in a country’s standard of living.

[T]he economic effects of policies are difficult to evaluate; long run effects are the most difficult. Engen and Skinner conclude, for example, that we really do not know the extent to which tax policy can change long run growth.

As set out previously,^[211] Australia’s pre-CGT experience provides much evidence of high levels of tax arbitrage and avoidance emanating from a CGT preference such as the CGT discount. Thus taxing capital gains at ordinary income tax rates would promote efficiency since it helps stem wasted resources on tax avoidance and evasion.

Under the CGT discount capital gains are not fully included as assessable income and thus capital losses are not generally fully deductible. If such losses were deductible against capital gains this would provide a greater incentive for people to save, invest, take risks and be entrepreneurial, as the government shares in the risk of such saving and investing.^[212] For example, a taxpayer in the top marginal tax rate of 48.5 per cent will virtually split the risk of investing in an asset subject to CGT with the government where capital losses are fully deductible against other capital gains. Removing the CGT discount permits greater deductibility of losses.

A case for fully taxing capital gains can also be made on the grounds that it builds the social capital^[213] of a society and thus enhances economic growth. Slemrod^[214] argued that a society has a stock of social capital and that trusting societies tend to have stronger incentives to innovate and accumulate both physical and human capital. Knack and Keefer^[215] tested the impact of these attitudes on the growth and investment rates of 29 countries, using a measure of trust and civil norms from world surveys in 1981 and 1990–91. They found that social capital variables exhibit a strong and significant positive relationship to economic growth.^[216] Trust is more correlated with per capita income in later years than early years showing that causation runs from trust to growth.^[217] Also, Hjerppe’s ^[218] survey of the literature found a close correlation between high levels of social capital and economic success. This also ties in to an extent with the Australian Productivity Commission’s^[219] analysis of underlying influences on economic growth.

Slemrod^[220] also found social capital possibly explained the most striking puzzle of public finance, the positive association with a country’s tax to GDP ratio and its level of affluence. He surmises that a high level of social capital and rule obedience contributes to affluence and facilitates a higher level of government activity than otherwise.^[221] He acknowledges other explanations for this puzzle; that is, rising income and urbanisation engender a greater demand for government services.^[222] He further notes that affluence is associated with demographic factors such as literacy and non-agricultural market activity and that these also require more government services.^[223] This suggests that higher levels of tax revenue would raise social capital.

Relevantly, a policy change lowers social capital if it reduces the incentive to be a good citizen or fails to increase the cost of not being one.^[224] Thus, ‘voluntary’^[225] taxes, such as Australia’s realisation CGT are risks to social capital. This tax is very ineffective as seen by the significant level of CGT exemptions such as the CGT discount.^[226] Additionally the risk of tax audit is low.^[227]

Given Australia’s extensive history of tax avoidance and loophole ridden tax legislation,^[228] one may surmise that the stock of social capital has dissipated. This is evident from an ATO survey of taxpayers' attitudes to income tax.^[229] The ATO refused to release the results claiming that it would not be in the public interest to do so,^[230] implying a very negative public attitude to income taxation, which lowers social capital.

As discussed previously, Australia’s CGT only produces modest levels of tax revenue, but CGT plays a greater role in that it prevents leakage from income tax. However, it is clear that taxing capital gains at normal tax rates would produce significantly greater levels of tax revenue.^[231] This would permit a reduction to income tax rates. As the following analysis suggests lower income tax rates may aid economic growth.

The following commentators have noted the positive economic effects of income tax base broadening and lowering personal income taxes. The International Monetary Fund (IMF)^[232] recommended reduction of the income tax rate graduations and broadening of the tax base by limiting deductions, exemptions, and other tax preferences. Additionally, Tanzi^[233] undertook an international comparison of individual income tax and economic growth, finding that a highly progressive income tax with many exemptions is highly detrimental to economic growth.

Indeed such a policy generally falls in line with international tax reform trends. Sandford^[234] illustrates this worldwide reduction of personal income tax rates made possible by the base broadening of the 1980s and beyond in the following table.

Top rates of central government personal income tax for 1976, 1986 and January 1992 for selected OECD countries

However, more recently there has been an upward movement of marginal individual income tax rates. In 2001 the top individual marginal income tax rate of 48.5 per cent in Australia was comparable to the top marginal income tax rates in other OECD countries.^[235]

Additionally, the following evidence suggests that a lower income tax rate from such base broadening may stimulate labour and human capital and thus aid economic growth. The Australian Productivity Commission^[236] highlighted the importance of human capital to economic growth. The Commission found that a society’s level of human capital had a close and tangible link with economic growth, along with other factors.^[237]

Furthermore, Boyer and Russell^[238] found that human capital is the most important determinant of wealth and income for most individuals—that is salary and wage earners.^[239] Yet they noted that most income tax analysis usually focuses on physical capital and labour supply.^[240] They asserted that in today’s knowledge-based economy the fundamental ingredients of growth are people and brain power.^[241] The manufacturing economy is becoming less and less important.^[242] Thus they question the provision of investment benefits in plant and equipment.^[243] On this basis, it appears that income tax rates for labour should be reduced to stimulate labour and human capital and thus aid economic growth.

There has been little research though on the influence of taxation on human capital. Heckman et al^[244] examined the impact of a switch from income to consumption taxation that exempts capital gains. They found that a consumption tax will be more pro capital but less favourable to human capital as it reduces the aggregate stock of human capital and the stock of human capital per worker for each skill group.^[245] Judd^[246] analysed optimal taxation in a dynamic model with human capital, finding that consumption tax proposals are biased against human capital as they deny expenses for human capital investments. This bias to physical capital breaches the neutrality goal of consumption taxes.^[247] Judd also noted that human capital might have a greater return than comparable financial assets.^[248] Education means that the risk of unemployment is low thus the price of risk attached to human capital is smaller than corporate equity.^[249] Judd found, though, that there is little empirical evidence about the critical determinants for the theory of optimal taxation of human capital.^[250]

However, there is considerable research on the impact on labour caused by tax rate changes. The responsiveness of labour to tax changes is known as the intratemporal elasticity of substitution.^[251] If a CGT preference leads to a tax increase for wage income and after tax wages are decreased, people may either increase or decrease the labour supply.^[252] A lower income may lead to consuming less leisure and supplying more labour. This is known as the income effect.^[253] This effect is offset, however, by lower wages making leisure cheaper.^[254] The latter is known as the substitution effect.^[255] This is defined as the desired ratio of goods to leisure changes in response to a change in the ratio of wages to the price of goods at a particular age, compensating for the income effect of the wage change.^[256] A large value implies that a decrease in after tax wages from a higher tax rate will lead to a large decrease in labour supply, and thus lower economic efficiency.^[257]

The research suggests that this responsiveness does not have a significant impact on the economy. Randolph and Rogers^[258] and Gravelle^[259] reviewed the empirical evidence on labour responses and found that the labour supply response, too, is probably small. Also, Goolsbee^[260] examined the responsiveness of taxable income to tax rate changes using data on several thousand corporate executives from 1991–95. The higher tax rates of 1993 led to a short run decline in taxable income since the responses came almost entirely from a large increase in exercise of stock options in anticipation of rate increases.^[261] There was no permanent effect.^[262] Executives without stock options, though, showed little responsiveness to the tax rate changes.^[263]

Sammartino and Weiner^[264] analysed the responses of taxpayers’ reported taxable incomes to tax rate increases in the 1990s. They acknowledged previous research that found that the tax rate cuts in 1986 caused a disproportionate increase in reported incomes for high income earners.^[265] They found that whilst the tax increases in 1993 resulted in a fall in reported incomes for high income taxpayers, especially for salary and wages, the changes were not permanent.^[266] By 1995 the decline in income for the highest income group was recouped.^[267]

Overall the removal of the CGT discount would permit a reduction to income tax rates and this may have a positive albeit small impact on economic growth. However, more research is needed to quantify the effect.

There are numerous arguments put forward in support of the full exemption or preferential treatment of capital gains. These include the impact of such a policy on: savings, investment, economic growth, lock in, risk taking, entrepreneurial activity, new ventures, equity financing, international competitiveness and foreign investment, as well as the impact of inflation. These issues will be dealt with in turn.

It is contended that a realisation CGT drives a wedge between current and future consumption and thus distorts the decision to save.^[268] Preferentially taxing capital gains reduces the size of this wedge and thus, it is argued, encourages saving.

Some commentators^[269] consider that the level of saving and investment is linked to capital formation and they assert that capital formation is the driving force behind economic growth. Indeed countries establish targets for capital accumulation.^[270] Consequently, it is argued that a CGT exemption or preference, such as the CGT discount, stimulates savings and thus spurs investment, capital accumulation and economic growth,^[271] as seen in the following diagram.

The government’s Ralph reforms appear to be at least partially inspired by the push to increase savings. The Treasurer and Minister of Finance stated that ‘raising national savings is the primary focus of the Government’s medium term fiscal strategy.’^[272] The Treasurer, Peter Costello, also stated that^[273] ‘with more saving we can finance more out of investment to grow the economy faster and produce jobs. Higher savings will raise the speed limits to growth.’ The Explanatory Memorandum^[274] accompanying the Ralph CGT reforms stated that ‘[t]he New Business Tax System will provide Australia with an internationally competitive business tax system that will create the environment for achieving higher economic growth, more jobs and improved savings’.

The influence of the CGT discount on savings, investment and capital accumulation on economic growth though is highly debatable and as the following analysis shows, there are weaknesses in every link of this chain of argument.

A lower CGT rate provides a higher after tax return on savings^[275] and this may increase private savings but it will also reduce government revenue. Government saving is important since the smaller the government debt the more funds are available for private investment, and the lower the cost of capital for business.^[276] Overall the impact on national savings of exempting or preferentially taxing capital gains will depend on whether a CGT preference results in an increase in private savings in excess of the increase in government debt. The following discussion explores this impact.

As noted above,the government’s rationale for the Ralph CGT preferences included the positive impact the reforms would have on savings. A CGT preference provides a greater after tax return as demonstrated by the following example that compares the returns on discounted (50 per cent CGT discount) and undiscounted growth shares held by an individual in the top marginal tax bracket. The shares are held for five years and appreciate at a constant 10 per cent per annum and pay no dividends.

The undiscounted shares will be worth $131 and the discounted shares $146 at the end of five years. Thus the after tax value of the discounted shares is 11.3 per cent more than the undiscounted shares’ value.

However, there are a number of reasons that suggest that a higher rate of return from a CGT discount will not have much of an impact on private savings. Firstly, realised capital gains comprise only 31 per cent of reported capital income.^[277] Many capital assets produce a significant part of their returns in the form of dividends, interest and rent. Additionally, a CGT discount would not greatly impact on tax exempt bodies or low taxed superannuation funds.^[278] Since non-resident investors in Australian shares are generally not subject to Australian CGT,^[279] the CGT discount will have no impact on these investors.^[280] Further, saving that is done for precautionary or liquidity reasons would be insensitive to the rate of return. As discussed later, lock in^[281] from a higher CGT rate will also increase the effective return from saving.^[282]

Further, the following research suggests that savings do not appear to be very responsive to an increase in the return of saving. As discussed previously,^[283] the responsiveness of certain economic parameters (such as savings or labour) have a substitution and income effect. Failing or lightly taxing capital gains appears to provide an incentive for people to increase their savings rather than consume (the substitution effect).^[284] Offsetting this though is the fact that people will save less given a higher rate of return to enable a higher level of current consumption whilst maintaining their future consumption needs (the income effect).^[285] This responsiveness is defined as the resulting percentage change in the ratio of future to current consumption, substitution effect, after compensating for the income effect. A large value makes it more likely that there are large efficiency gains from either switching from an income tax to a consumption tax,^[286] or by exempting capital gains from taxation. As these effects work in opposite directions it is not clear whether a CGT exemption would actually spur savings.

Clearly the introduction of CGT in Australia in 1985 was not associated with a fall in savings. Rather, during the 1990s the ratio of net national saving to GDP rose from 2 per cent to 2.8 per cent.^[287]

International researchers have not found a strong responsiveness of savings to tax rate changes. Burman’s^[288] research suggests that the impact of taxation on savings is minimal. He noted, however, that there is a degree of ambiguity in the theory.^[289] Also, a literature survey by Randolph and Rogers^[290] concluded that empirical studies of this elasticity provide little information to base a confident judgement on the responsiveness of savings to tax rate changes, but they found existing estimates suggest that the value is small. Similarly, Gravelle^[291] found that the empirical research is limited for such savings responses and faces greater challenges. Gravelle found that, overall, the research suggests small effects of uncertain sign for savings responses to tax rate changes.

Additionally, Kay and King^[292] asserted that savings are not very responsive to the rate of return, as the private sector savings did not exhibit volatility during the 1970s and 1980s when large changes in real interest rates occurred in the United Kingdom and where real interest rates were sometimes negative after tax.

Gann^[293] found that empirical studies have been unable to show that a reduction in income tax rates will result in an increase in saving. Also, she notes that gross private savings have been constant, staying between 16–17 per cent of Gross National Product (GNP) since 1951.^[294] Thus, she notes that some economists argue that government savings via surpluses in periods of high employment are preferable to stimulating private savings via tax preferences.^[295]

Gordano^[296] considered that saving and borrowing were not heavily influenced by after tax returns and costs, finding that Americans gave up current for future consumption very grudgingly. Since the early 1970s to mid 1980s real consumer spending has not deviated much from its trend.^[297] Thus he considered that tax has only a small influence.^[298]

Further, Boyer and Russell^[299] noted that during the 1980s, tax rates were reduced and new savings incentives introduced but the rate of savings in the United States still fell. Also, they noted that Japan had the least consumption tax but the highest rate of savings and GDP growth in the 1980s^[300]. Further, a 1986 OECD study^[301] of 23 OECD countries compared the rate of national savings as a percentage of GDP to each country's use of consumption tax, and found no correlation.

Additionally, studies by Sears, Roebuck and Co Economics Department^[302] showed that the lack of savings has more to do with demographics of the population than any other item. In the 1980s the United States had a young population and, as it ages, savings will increase.^[303] Also, Tanzi^[304] undertook an international comparison of the individual income tax and economic growth. He found that savings are not responsive to tax changes.^[305] Krever and Brooks noted that during 1978–86 private savings in the United States declined to the lowest level since World War II yet the CGT rate was reduced from 35 per cent to 28 per cent and then to 20 per cent.^[306]

The Canadian government^[307] estimated that the exemption of capital gains from tax raises the after tax rate of return on total private savings by only a small amount: one quarter of a percentage point.

Whilst a CGT discount reduces tax revenue and thus lowers public savings, some economists^[308] believe that such a policy would actually have the opposite impact. It is suggested that a CGT cut would increase the number of CGT realisations so as to raise tax revenue and government savings.^[309]

The responsiveness of asset realisations to CGT cuts is known as ‘the elasticity’.^[310] If the elasticity is zero a CGT cut will not change taxpayer behaviour.^[311] If the elasticity is one or more, a CGT rate cut would be self financing^[312] since tax revenues would be maintained as a result of a greater number of realisations. Given that shares and other marketable securities provide the vast majority of reported capital gains in Australia,^[313] research by Feldstein et al^[314] is highly relevant. This research was based on cross-sections of tax returns and found that capital gains on shares were very responsive to tax changes and a rate cut would not result in any loss of tax revenue.^[315]

Reynolds made a similar finding on research provided to the Ralph Report in relation to lowering Australia’s CGT rate. Reynolds^[316] found that a cut in the CGT rate would increase the realisation of assets. The elasticities of realisations were estimated at 1.7 in the short-term and 0.9 in the long term, meaning that the CGT rate reductions would be revenue positive.^[317]

A lower CGT rate may produce extra revenue in the short run since appreciating assets will be more attractive to investors.^[318] Demand will drive up the price of such assets and this will lead to greater capital gains.^[319] However, there are a number of reasons why elasticity will not be high. As noted above, a CGT rate cut will have little impact on tax exempt assets and investors such as exempt bodies, exempt foreign investors and low taxed superannuation funds. Also, Burman^[320] noted that the long-term price of capital assets is based on their replacement costs. Such costs will not be affected by a CGT cut thus any short-term gains will be at the expense of longer-term gains.^[321] Thus Burman concluded that this only has a transitory impact on tax revenue.^[322] Further, a CGT preference will lead to the transformation of other types of income into capital gains.^[323] The elasticity measure ignores these losses in tax revenue.^[324]

Additionally, Burman^[325] noted that the models used by economists to determine the ways investors realise assets imply that the elasticity depends on how taxpayers value their portfolios and on the diversity of their attitudes. Thus, if realisations are highly elastic, investors should have similar expectations. However, he points to the high CGT rates in the United States in 1996 and the plethora of realisations.^[326] Another contributing issue raises the question of who buys the assets when everyone is selling.^[327] This diversity of investor attitudes implies a lack of sensitivity to rate changes.^[328]

Also, some assets such as real estate are illiquid and have high transaction costs, such as stamp duty and selling fees. Such assets would be less sensitive to CGT rates.

Thus, it is not surprising that other research, based on time-series studies, suggests that elasticities are quite small.^[329] Empirical research by Burman and Randolph^[330] on variances in tax rates among states in the United States sheds some light on these conflicting studies of the elasticity of asset realisations. The study found that the permanent effect of a tax rate cut on individuals’ responses was very small and close to zero.^[331] Individuals’ short-term responses, though, varied greatly in response to changes in tax rates. Although, Burman^[332] noted that this does not answer the impact on tax revenues and concluded that it is unlikely that a permanent elasticity would be large enough to be self financing.^[333] To achieve this unlikely scenario the rate cut would have to be small since elasticity increases with tax rates.^[334]

The introduction of the CGT discount appears to illustrate the deficiencies with Reynold’s approach. In 1999 the following evidence about the impact of a CGT rate cut^[3]35 on tax revenue collections was provided to the Australian Senate Finance and Public Administration References Committee Inquiry into the Business Taxation Reform:^[335]

Capital gains tax elasticity estimates provided to the Australian Senate Inquiry

The Ralph Report^[336] expected that a cut in the CGT rate would increase the realisation of assets and thus be revenue positive.^[337] Gravelle^[338], a senior economic specialist in the Congressional Research Service in Washington, concluded, however, that the Ralph Report’s long run elasticity figure was too high. She estimated that the long run elasticity was around 0.2 and the short run elasticity was about 0.6. Auerbach^[339] had similar findings. Both Gravelle and Auerbach considered that Reynold’s findings were inadequate, as he relied on methodologies containing significant flaws and was unable to distinguish between transitory responses to temporary tax changes, short run responses to permanent tax cuts and long run permanent effects.^[340]

Indeed, the CGT discount has proven to be very costly to tax revenue (see Treasury estimates and projections illustrated in Table 2, above, which shows revenue losses of $1.8 billion per annum from 2001–02 through to 2004–05). ^[341]

Table 1 (above) sets out net capital gains and CGT revenue before and after the September 1999 Ralph CGT reforms. The tax revenue lost on the CGT discount of $1.3 billion^[342] in 2000–01 is very substantial compared to the CGT revenue of $4.4 billion. Further, CGT realisations and CGT collections have actually fallen since the Ralph Reforms. Thus it can be inferred that the CGT discount has not had a positive affect on CGT revenue. This may, however, be largely due to other factors such as the decline in the stock market.

However, the average number of daily transactions on the Australian stock exchange appears to have increased markedly following the introduction of the CGT discount. As Evans^[343] noted, prior to the CGT discount, the average number of daily share transactions in the quarters ended 30 September 1998 (27 190 transactions), 31 December 1998 (28 390 transactions), 31 March 1999 (35 290 transactions), 30 June 1999 (40 250 transactions) and 30 September 1999 (39 300 transactions). Following the CGT discount, the average number of daily share transactions increased, in the quarters ended 31 December 1999 (47 260 transactions), 31 March 2000 (75 870 transactions), 30 June 2000 (57 130 transactions), 30 September

This increase for March and April 2000, though, would appear to be largely attributed to the height of the tech boom.^[345] Thus it appears difficult to filter out the impact of the CGT discount. Notably, the CGT discount is generally only available for assets that are held for 12 months^[346] and this limits its availability. Other factors clearly impact on share turnover. For example, the average number of daily share transactions were 36 per cent greater in the month of August 2002 compared to the month August 2003.^[347] Given that the CGT discount had been present for about four years it is logical that other factors, such as market supply and demand, drove this increase. Notably, the largest trading day up to 29 August 2003, with $6982 million worth of share trades, occurred prior to the introduction of the CGT discount (on 22 June 1998).^[348] Again, this suggests that the CGT discount may not have had a great impact on stock market turnover.

It is unlikely that private savings will be greatly affected by the CGT discount given the limited responsiveness of savings to the rate of return. The CGT discount has reduced the level of government savings and it is also unlikely that private savings would have increased sufficiently to make up this shortfall of about $1.8 billion per annum.^[349] Overall national savings would appear to decline under the CGT discount.

Schmidt^[350] and Chaudhri et al^[351] both found that there does not appear to be a clear link between Australian savings and investment. Further, the Australian Productivity Commission^[352] observed that investment expenditure is closely related to macroeconomic conditions, rather than savings. Indeed the Australian Bureau of Statistics^[353] noted that between 1990–92, real gross capital formation per capita fell by more than 6 per cent and that this was associated with the recession experienced by Australia in the late 1980s and early 1990s. Additionally, since savings are affected by international capital flows, an increase in domestic savings may not increase domestic investment.^[354]

Similarly, international researchers, Boyer and Russell^[355] did not find a relationship between savings and investment. They found that investment decisions are made on projected future consumption, and that investment tracks consumption.^[356] They found that investment has been increasing in the United States in recent years, although savings had declined.^[357]

Even assuming that the CGT discount could provide a higher level of investment it can be queried whether it is appropriate to give the preference to all types of assets, productive and non-productive?^[358] Thus, a CGT discount, if justified at all, should be targetted to productive assets.

Research indicates that there does not appear to be any clear link between savings and economic growth. A study by Schmidt^[359] examined the long-term and short-term relationship between Australia’s savings, investment rates and economic growth. He found that in regard to the various neoclassical growth models that a policy directed at increasing saving would not change national investment and thus not affect economic growth.^[360]

Another study, by Chaudhri and Wilson,^[361] used a growth model to examine the relationship between Australian savings, investment and productivity growth for the periods 1861–1900 and 1949–90. They found that an increase in savings was neither necessary, nor a necessary requirement, for economic growth in a small open economy like Australia.^[362] Australian savings have been well below investment rates throughout most of Australia’s economic experience, with foreign capital funding capital accumulation.^[363]

The Australian Productivity Commission^[364] found that Australia’s high rate of economic growth in the 1990s could not be explained by a high level of savings but rather by an increase in multifactor productivity stemming from a range of factors. These include an increase in the adoption of new technologies, greater use of innovation and research and development, improvements in human capital, organisational change, new management techniques, enterprise bargaining, reallocation of resources, greater specialisation, greater openness and competition in the Australian economy and changes to business expectations about reliance on government support.^[365]

International research supports this view. Gravelle^[366] found that there is no automatic argument that greater savings would spur economic growth. An increase in savings reduces current consumption and increases future consumption when these savings are consumed.^[367] This must damage current economic growth given the reduced level of current consumption. This shifts consumption and hence welfare to future generations.^[368]

(b) Capital Gains Tax and Investment and Capital Accumulation and Economic Growth

It is further argued^[369] that CGT reduces the level of investment and capital accumulation since it increases the cost of capital and thus capital gains should be concessionally taxed. The American Council for Capital Formation^[370] found that the 1997 CGT cuts in the United States from 28 per cent to 20 per cent reduced the net cost of capital for new investment by around 3 per cent. The Council argued that United States CGT rates are higher than other industrial countries and that savings rates and investment in recent decades was low.^[371] Additionally, Goolsbee^[372] found that tax significantly affects prices and investment and that conventional research results understated this by a factor of four. Using panel data on different types of capital equipment, he found a measurement error in the tax component of the cost of capital.^[373] This measurement error stems from the impact of expected future tax reforms, tax loss asymmetries, inflation and the complexity of tax laws.^[374]

These concerns must be balanced firstly against the impact on government debt of the CGT discount.^[375] Secondly, the CGT discount offers no benefit for productive depreciating assets that are exempt from CGT.^[376] Thirdly, the CGT discount reduces the value of deductions for investments that fail. Logically, this may deter investment. Fourthly, much of the capital provided for investment is provided by low taxed superannuation funds^[377] and CGT exempt foreign investors^[378] and is thus not greatly affected. Fifthly, large corporations would not be greatly impacted by an increase in the cost of capital that may make share issues unattractive since they can borrow funds or rely on retained earnings.^[379] Companies would also not be greatly affected given the low company income tax of 30 per cent. Certain individuals though would be affected given the high top marginal income tax rate of 48.5 per cent.

Finally, the introduction of the Australian CGT in 1985 did not appear to result in any discernible slump in Australian investment and capital accumulation.^[380] The Australian Bureau of Statistics^[381] found that between 1992–93 and

Other international research suggests that a CGT preference has a minor impact on investment and capital accumulation. Hulten^[382] concluded that a CGT rate reduction would not have a great impact on the cost of capital. Tax changes may affect marginal investment, but investment that embraces new technology will not be marginally profitable.^[383]

Overall, the CGT discount does not appear to greatly reduce the cost of capital. Such a policy is likely to have only a marginal impact on investment and capital accumulation.

A study by Chaudhri and Wilson^[384] found an interdependent relationship between Australian investment and productivity growth for the period 1949–90. They noted that the interactions between investment, GDP and productivity growth were complex and evolving.^[385] Other public finance specialists, such as Feldstein,^[386] assert that capital accumulation plays a central role in the growth of every economy, and thus is a major focus of economic theory and empirical research. Gann^[387] also found a broad consensus that higher levels of capital formation are desirable.

The American Council for Capital Formation^[388] referred to research by Bradford and Summers that found investment in equipment is perhaps the single most important factor in economic growth, with every 1 per cent of GDP invested in equipment being associated with an increase in GDP growth of one-third of a per cent. The Council^[389] also argued that the slow economic growth in the United States in the last twenty years is partly attributed to low investment.^[390] They pointed to a recent study by the World Bank^[391] that suggested countries with high levels of investment grow faster. The research referred to fifteen countries in the study that compared average annual GDP growth for 1974–93 and savings and investment. There was a general correlation between these parameters, with some sharp exceptions.^[392] Hong Kong had double the GDP growth of Japan, yet its level of savings and investment was only slightly higher than Japan’s.^[393] South Korea had high saving and investment, but not the highest GDP growth, whereas Denmark had low saving and investment, but not the lowest, and had the lowest GDP growth.^[394] This comparison, though, was unusual in that it did not refer to comparable countries, that is, OECD countries. The Council also referred to Jorgenson’s findings that increases in capital stock had the strongest impact on the growth in output.^[395]

Other evidence suggests, however, that capital accumulation may not have such a strong impact on economic growth. Firstly, depreciable assets will not be affected by the CGT discount given the exemption from CGT.^[396] Further, the Australian Productivity Commission found that investment expenditure is not a direct contributor to productivity growth, with productivity gains driving growth.^[397] This is also borne out by recent OECD^[398] data that shows that capital accumulation is becoming less important as a factor in Australia’s growth given the increasing dominance of services as a proportion of GDP. In 1987 services accounted for 64.9 per cent of GDP and 70.6 per cent in 1997.^[399] Over the same period industry declined from 31 per cent to 26.2 per cent and agriculture declined from 4.1 per cent to 3.2 per cent.^[400] Notably this trend is present in virtually every OECD country.^[401]

Other international commentators have found that investment and capital accumulation have a relatively small impact on economic growth. Pechman^[402] disagreed with Feldstein et al’s analysis above, finding that capital accumulation is not as significant. He estimated a 1 per cent increase in the ratio of investment to GNP will only generate a 0.2 per cent increase in productivity.^[403] Although Fazzari^[404] noted that most quantitative analyses assume a 1 per cent increase in capital stock will increase output by 0.3 per cent, he felt this only has a small transitory impact. It has no permanent effect, as the level of output permanently changes, not the growth rate of output.^[405] Also, Fazzari^[406] placed the impact of a CGT rate cut on economic growth in the context of economic modelling terms. He asserted that when one multiplies three numbers substantially less than one, then this results in a very small number.^[407]

Relevantly, Gann^[408] found that productivity growth declined in the United States in the late 1970s and early 1980s, even though investment in non-residential fixed capital as a percentage of GNP remained strong throughout the 1970s.

Hulten^[409] similarly concluded that multifactor productivity, not capital formation, accounted for nearly all the postwar growth in the United States. He found that variations in economic growth are closely tied to multifactor productivity.^[410] From 1948–73 the United States business sector grew at 3.6 per cent and from 1974–81 it fell to 2.3 per cent.^[411] The decline was almost all due to the fall of multifactor productivity.^[412] Measuring multifactor productivity is difficult, as it is a combination of factors.^[413] These include technical and organisational change, changes in the quality and utilisation rates of capital and labour, changes in allocational efficiency and pure measurement error.^[414]

Bosworth^[415] also considered that capital formation had a limited impact. The United States productivity slowdown in the late 1970s was only marginally affected, by only 0.1 to 0.2 percentage points, by differing rates of change in the capital labour ratio.^[416] Gravelle^[417] noted that the slowdown in productivity growth in the 1970s has not been easily explained, but seemed unrelated to tax policies or capital accumulation rates.

The following comparison of 29 developed countries does not show any link between CGT policy and economic growth.^[418]

If anything, this table shows that the absence of a CGT is associated with weak rather than strong economic growth. Some of the highest GDP growth rate countries such as Ireland, Iceland and Poland all have a CGT. Czech Republic and Italy have only minimal capital gains taxes and yet are very poor performers on GDP growth. New Zealand has no CGT and has poor growth. Further, Burman found that in the United States, a dozen changes to CGT rates in the past 45 years has shown no perceptible affect on economic growth.^[419]

Overall there appears to be a relationship between investment, capital accumulation and economic growth. The Australian and international research though indicates that investment may not have a strong influence over economic growth in Australia.

The CGT discount is a very blunt instrument to raise savings and such a policy appears to reduce Australian national savings. Even if it raised national savings, there is no clear link between Australian savings and investment, nor between Australian savings and economic growth.

The CGT discount reduces the cost of capital but the impact on economic growth appears to be small. Whilst the cost of capital is related to capital formation, the introduction of the Australian CGT did not appear to have dampened capital formation. Capital formation does contribute to Australia’s growth but it appears to be subordinate to other factors such as productivity gains.

Overall, the impact of a CGT discount on economic growth does not appear to be significant.

Since a CGT liability usually only arises upon the sale of an asset, this encourages taxpayers to hold onto assets for longer than otherwise planned. Thus investors may be unwilling to sell assets in response to new information. This is known as the lock in effect.^[420] This results in a misallocation of capital among industries and firms.^[421] It distorts the demand for portfolio assets, alters the mix of financial assets that firms supply, the ratio of debt to equity and the allocation of risk bearing in the economy.^[422] Capital may be retained in less productive or inefficient investments.^[423]

However, there are a number of reasons why the lock in effect would not have a significant economic impact. The vast majority of reported capital gains arise from shares^[424] and it is apparent that major players in the Australian share market, lower taxed superannuation funds^[425] and foreign investors^[426] are not greatly affected by Australia’s CGT. Such investors would sell quickly in response to poor company performance.^[427] Additionally, lock in does not apply to CGT exemptions such as pre CGT assets, main residences and depreciating assets.

Further, the Ralph Report must not have thought lock in was a problem for companies since the CGT discount was not extended to companies.^[428] Also, the tax deferral advantage of CGT may have a reverse lock in effect.^[429] That is, an alternative investment may appreciate faster and thus provide an even greater benefit from CGT deferral.^[430] Thus an investor may sell too soon and pay taxes now rather than in the future.^[431] Additionally, the CGT discount may encourage another type of lock in given the requirement for a 12-month holding period.^[432]

Lock in would also not have a great impact on many taxpayers since investors that receive capital gains generally own a number of assets.^[433] With a large portfolio of assets investors will have capital losses and gains, thus capital loss assets can offset any gains. An investor can then simply buy back the loss asset or buy a replacement asset to avoid any real economic loss. Alternatively, an investor can reduce lock in by buying other assets whose returns are negatively correlated with the asset subject to lock in.^[434]

There may in fact be some benefit from lock in. An individual’s trading may be made on the basis of information on the future potential of an investment or be made on ignorant speculation.^[435] Clearly, ignorant speculation does not improve the efficiency of the market and history indicates that investors overreact to short-term information on profits, mineral discoveries or new technologies.^[436] This was evident in the dot com share boom in Australia and overseas.^[437] Thus there may be some benefit gained from lock in by reducing ignorant speculation and short-term overreactions, as it stabilises investments.^[438] Also, people may hold investments rather than consuming, this promotes investment. Further, taxing capital gains and allowing deductions for capital losses may also increase the mobility of capital in recessions, and thus enable the transfer of capital to more efficient uses.^[439]

Krever and Brooks^[440] point to a number of situations where the lock in effect will not apply to asset holders. For example, where the taxpayer has no choice but to sell assets for business considerations or through forced disposals.^[441] Also, lock in will not be much of a problem for assets having a short life such as leases, patents, minerals deposits and wasting assets.^[442] Additionally, it will only marginally apply to assets having low capital gains.^[443] Relevantly, the average capital gain in 2000–01 was only $4726.^[444] There will also be negligible impact on individuals having low marginal income tax rates.^[445]

Overall, lock in does not appear to be a great problem. If lock in was considered to be a problem then death should not be exempted from CGT. Exempting CGT on death provides one of the primary reasons for lock in since it enables the elimination of all CGT liability.

As Burman^[446] noted, some argue that CGT results in the double taxation of corporate income since corporate income is subject to company income tax and dividends are taxable at the shareholder level. This results in a misallocation of capital to the corporate sector and results in high levels of corporate debt.^[447] Ideally, corporate and personal income tax systems should be fully integrated since this enhances economic efficiency.^[448] This prevents the double taxation of income at the company level and of the individuals who are the ultimate owners of the business entity. Under full integration the income of all entities would be attributed to individual taxpayers who are the ultimate owners of the company.^[449] The income tax would be payable by these individuals at their marginal rate of income tax.^[450] Given a number of practical considerations no countries have implemented a full integration regime.^[451]

However, this argument does not apply since Australia utilises a partial form of integration known as the dividend imputation system.^[452] This provides tax relief to domestic shareholders for company tax paid. The individual shareholder includes the sum of the dividend plus the tax credit, known as the grossed up amount as assessable income.^[453] The taxpayer’s marginal tax rate is then applied to the grossed up amount and this is reduced by the credit for the corporate tax paid. If the taxpayer’s marginal income tax rate is greater than the company tax rate of 30 per cent (assuming that the dividend is fully franked) additional tax is payable, if less a refund is due. Thus a taxpayer on the top marginal income tax rate of 48.5 per cent would pay an extra 18.5 per cent tax on a fully franked dividend.

The CGT discount though may distort corporate equity markets.^[454] Under this policy companies favour retained earnings since shareholders prefer CGT advantaged capital growth.^[455] Companies have less reason to pay out dividends and are less reliant on new share issues to finance growth.^[456] This means that investment markets are unable to allocate capital to those firms that produce the highest returns.^[457] Notably, the Ralph Report did not extend the CGT discount to companies.

Some commentators^[458] argue that real capital gains should only be taxed and that this enhances economic efficiency. Thus a CGT discount is required to offset the impact of inflation. As discussed previously,^[459] inflation provides a comparative advantage to capital gains compared to other forms of capital income. Consequently, any indexation of capital gains would require the indexation of other types of capital income and the income tax system. Full indexation though is highly unlikely given that no country appears to have embraced such a policy. Consequently the inflation argument for a CGT preference such as the CGT discount lacks merit.

As noted above, it is argued that a lower CGT rate will decrease the cost of capital for business. Thus it is argued that this may encourage risk taking,^[460] equity financing,^[461] entrepreneurial activity and new business start ups.^[462] In particular fully taxing capital gains would appear to affect businesses that rely on the equity market such as public utilities that need large amounts of capital and for high growth companies that need to frequently raise equity capital.^[463] Small growth companies have insufficient retained earnings and limited borrowing capacity to raise funds would also be affected.^[464] Thus it is argued the higher the cost of capital will reduce investors’ appetite for risk and thus some of the most dynamic and productive sectors of the economy may be hindered.^[465]

Relevantly, Gompers and Lerner^[466] examined the forces affecting fundraising by independent venture capital organisations from 1972 to 1994, finding that the surge in venture capital in the 1980s was a product of demand of venture capital by entrepreneurs, stimulated by a growing economy, technological change and CGT rate cuts.

These arguments though may be overstated. The finding by Gompers and Lerner has been subject to some criticism.^[467] Blair casts doubts since the paper was based only on a small sample size of 17 relevant observations.^[468] This does not have much power to reject the null hypothesis that the CGT rate does not matter to risk taking.^[469]

Further, Hellman asserted that given that an entrepreneur would not cash out business gains until a fairly distant time down the track, and that the investment or business returns are highly uncertain, then any changes to the current CGT rates must have little influence.^[470] Also, typically entrepreneurs realise their capital gains at a very slow rate, using various techniques to delay payment of taxes.^[471] It is unclear as to how entrepreneurs look at the current CGT rates as a predictor of future tax liability. Complaints about high CGT are likely to stem from successful entrepreneurs trying to avoid tax rather than entrepreneurs wanting to start a new project.^[472] Also, Poterba found that there is no empirical evidence that a CGT preference would stimulate entrepreneurial activity and venture capital.^[473]

Additionally, as McGee^[474] noted, a CGT preference provides encouragement to existing firms rather than new firms. New firms will have no or few capital gains since capital gains accumulate over time.^[475] Whereas established firms will enjoy capital growth from goodwill, real estate, investments and other business assets.

Also, since low taxed superannuation funds^[476] and exempt foreign investors^[477] are major contributors to equity financing this may not appear to be a significant issue since a CGT rate cut will have little impact on their after tax rate of return (as noted previously). Importantly, entrepreneurs largely provide human capital rather than financial capital to their business ventures and rely on outside investors for financial capital.^[478] Thus the current CGT policy may have little impact on many entrepreneurs.^[479]

Further, the CGT discount poorly targets entrepreneurship as it applies to all types of assets. For example the CGT discount applies to unproductive assets such as holiday homes, hobby farms, antiques, artwork and yachts. It does not apply to productive assets such as depreciating assets.^[480] Commentators have noted^[481] that a CGT preferences or exemption encourages low risk investments such as real estate and blue chip shares.

As noted above, if capital gains are not treated as income then logically capital losses will not be deductible. This may discourage risk taking. Indeed, Kaplow^[482] refers to a study of risk taking done by Domar and Musgrave^[483] that shows that an income tax in which investment losses are deductible without limit, increases risk taking.

Fully taxing capital gains would not have much impact on many large companies since they can borrow funds or use retained earnings.^[484]

Between 1985 and 1998, Australia’s CGT applied normal rates of income tax but this did not appear to have affected risk taking as seen by the steady increase in share and business assets since 1985.^[485]

Even assuming that a CGT preference will encourage risk taking there is no evidence that demonstrates risky rather than safe investment should be supported. Economist Sandmo surveyed the literature and found that ‘It seems reasonable to conclude that the few studies which have been made of the optimum taxation of risky assets cannot provide any a priori foundation for a recommendation that risky assets be taxed at higher or lower rates than safe ones’.^[486]

Overall, the evidence suggests that a CGT discount would not greatly discourage risk taking.

Since 1788, Australia’s small open economy has relied extensively on international trade and capital links.^[487] In this context, it is asserted^[488] that a CGT preference is necessary to maintain Australia’s international competitiveness.

This argument would appear to have merit if Australia imposed a harsher tax regime than other OECD countries. However, all other OECD countries, with the exception of New Zealand, impose CGT it would seem difficult to see how CGT could damage our competitiveness unless Australia imposed significantly higher CGT rates. Relevantly, Evans noted that prior to the 1999 CGT discount Australia’s CGT regime imposed higher levels of CGT than other OECD countries.^[489]

However, Australia also competes with non-OECD countries but does that mean it is appropriate to import the CGT policy of the lowest CGT country. Under this approach, the personal income tax and company tax rates must also be reduced to maintain our competitiveness. Also, why preferentially tax capital gains and not other types of capital income? Such approaches lower government expenditure and services, and thus undermine important institutions such as the education, health, defence and justice systems.

Further, a tax policy should promote competitiveness and this is best achieved through a competitive domestic economy.^[490] Thus a neutral tax policy that taxes capital gains at ordinary income tax rates would appear to allow market forces to better allocate resources.

Overall, the arguments run both ways and there does not appear to be a compelling case for preferentially taxing capital gains on the basis of international competitiveness. Although, more economic research is needed to ascertain the affects of discounted CGT rates on international competitiveness.

Further, it is argued^[491] that a CGT preference is necessary to attract foreign investment. Importantly, research suggests that inward foreign investment enhances economic growth.^[492] Given Australia’s strong reliance on foreign investment a harsher CGT may be costly. This reliance was noted in an empirical study from 1861 to 1900 and from 1949 to 1990 by Chaudhri and Wilson.^[493] The study found that Australian savings were well below investment rates through most of Australia’s economic experience and noted that foreign capital provided additional funding for capital accumulation.^[494] Also, in 2001–02, the total foreign investment inflow into Australia amounted to $72.5 billion.^[495] This amounted to a significant sum being 10 per cent of GDP in 2001–02.^[496]

However, a number of factors may reduce the impact of CGT on inward foreign investment. Firstly, CGT generally only applies to a limited number of foreign investors having permanent establishments and that acquire assets such as Australian real estate and shareholdings of greater than 10 per cent in Australian companies.^[497]

Secondly, the introduction of the Australian CGT in 1985 did not correspond with any fall in foreign investment. From 1991 to 2000, foreign direct investment in Australia increased from 5 per cent to 6 per cent of total Australian assets.^[498]

Thirdly, Gravelle^[499] noted other limits in the responsiveness of international capital flows. The mobility of capital may be restricted since investors do not see investments in different countries as perfect substitutes.^[500] Also, as Gravelle noted, investment flows require changes in trade flows. Additionally the tax revenue from attracting foreign capital may be limited as debt financed capital is subject to a negative income tax and investment subsidies attract both debt and equity.^[501]

Overall, it appears that the CGT discount provides some limited encouragement for foreign investment and this may aid economic growth. This must be balanced though against the economic costs from the lack of neutrality and the loss of CGT revenue. The problems of such a CGT preference is evident in the move by OECD countries to remove harmful tax preferences and to retain the taxation of capital income.^[502]

A CGT rate set at the same level that applies to other types of capital income would provide a more neutral treatment for capital income than the current CGT. This would lead to a better and more productive use of resources. Certainly there is considerable evidence of over investment in the building and housing industry. However, the economic cost of this misallocation of resources is unknown.

Additionally, the removal of the CGT discount would help close a major exemption in the income tax system as seen by the massive levels of discounted capital gains. This would restrict the opportunities for tax avoidance and minimisation. The removal of the CGT discount offers greater deductibility for capital losses and enhances social capital. Additionally, there appears to be some economic benefit from its great revenue raising ability that would enable the lowering of tax rates.

There are, however, many opposing economic arguments for the CGT discount. Overall the evidence suggests that a CGT levied at normal income tax rates though would not greatly hinder savings, investment and economic growth, nor would lock in be likely to have a significant impact. The inflation argument for preferential CGT treatment is baseless given that indexation does not apply to other types of capital income and to the income tax system.

The arguments based on CGT hindering corporate equity lack merit. Also, the CGT discount does not appear to greatly affect risk taking. Further, it is not readily apparent that the removal of the CGT discount would seriously damage international competitiveness or significantly discourage inward foreign investment.

Whilst the economic arguments run both ways, overall the arguments for taxing capital gains at ordinary income tax rates appear to be more compelling. More economic research is greatly needed though to quantify the economic benefits and costs of exempting, preferentially taxing and comprehensively taxing realised capital gains.

The reforms related to the CGT discount do not appear to hinder economic efficiency. Indexation for capital gains cannot be supported unless indexation is similarly extended across the income tax system. The preferential CGT treatment for averaging is limited to a few taxpayers and thus this leads to distortions and a loss of neutrality.

VII SIMPLICITY

There is only limited research into Australian CGT compliance costs but the data generally shows that compliance costs are high. The first study by Pope et al^[503] of federal compliance costs in the late 1980s to early 1990s found compliance costs to be significant. The study estimated personal income tax compliance costs of $3642 million or 9.2 per cent of personal income tax revenue in 1990–91. Corporate income tax compliance costs were estimated at $3246 million or 22.9 per cent of corporate income tax revenue^[504]. The data though did not separately identify CGT compliance costs.

A study by Wallschutzky and Gibson^[505] in a qualitative approach, examined 12 small businesses and their tax compliance costs. The study was conducted over a year from November 1991 to November 1992. They found few taxpayer problems complying with Commonwealth taxes, apart from sales tax.^[506] They also noted however, that CGT was a longer-term problem given the poor record keeping of small business.^[507] Nevertheless, the report concluded that compliance costs were not significant.^[508]

Following a study^[509] to examine the incremental costs of compliance, Evans et al^[510] were further commissioned by the Australian Taxation Office to estimate compliance costs for Australian taxpayers in respect of federal tax in the 1994–95 tax year. Using a large mail survey, the social compliance costs were estimated at $10.4 billion or 11.9 per cent of all tax revenue and 2.29 per cent of GDP.^[511] After taking into account the tax deductibility benefit and cash flow benefits from tax deferral, compliance costs fell to $6.2 billion, or 7 per cent of tax revenue and 1.36 per cent of GDP.^[512] Two and one half million business taxpayers incurred 75 per cent of the compliance costs and 7.3 million individual taxpayers incurred the other 25 per cent of costs.^[513]

The survey also found that only 1.2 per cent of 1528 personal tax (non-business) respondents received CGT income in 1994–95 tax year.^[514] Of these, 583 used paid tax practitioners but only 4.5 per cent used a practitioner because of a need for CGT advice.^[515] Rather, personal taxpayers mainly sought practitioners so as to prepare the income tax return.^[516] Personal taxpayers spent 18 minutes on average per annum in maintaining CGT records, which was only 2 per cent of time spent on tax compliance.^[517]

The survey found that business taxpayers generally had little exposure to CGT.^[518] There were 2462 business taxpayer respondents with 3.5 per cent having frequent CGT issues, that is four or more times a year.^[519] Another 19.6 per cent of business taxpayers had occasional CGT issues (1–3 times per week).^[520] Others never dealt with CGT (48.4 per cent), or failed to respond to the question (28.5 per cent).^[521] Business taxpayers spent four hours per annum on average on CGT matters out of a total of 90.7 hours.^[522] Only 4 percent of their time was spent on CGT.^[523] Thus CGT was not a significant compliance cost, estimated as costing $155 million out of $4.6 billion in total business compliance costs.^[524] Significantly, when compared to CGT revenue of $994 million, compliance costs were high, being 15.59 per cent of CGT revenue.^[525]

The compliance costs for CGT and other federal taxes are set out in the following table:^[526]

The table demonstrates that CGT compliance costs are generally significantly more expensive than other taxes when compared on a percentage of revenue yield basis.

Evans et al^[528] were further commissioned by the Australian Taxation Office to furnish annual updates of compliance costs for 1997–98, 1998–99 and 1999–2000 based on random stratified samples of over 3000 taxpayers. The 1997–98 figures have been produced and indicate significant changes to CGT compliance costs for business taxpayers compared to 1994–95.^[529]

The annual internal labour time spent on CGT issues rose from two hours to 27 hours.^[530] This meant that CGT internal compliance time increased from 4 per cent of all internal time to 13 per cent.^[531] It was estimated that CGT compliance costs for business taxpayers had risen from 3.3 per cent of all business compliance costs to 10.2 per cent.^[532] The data also suggested that the CGT compliance costs as a ratio to CGT revenue yield remained high at 13.1 per cent.^[533] Notably over this time, the number of business taxpayers receiving capital gains increased from 290 380 in 1994–95 to 762 679 in 1997–98.^[534] The increase in compliance costs may be attributable to the large increase in business CGT payers or it may be due to CGT changes such as the new business exemptions,^[535] rollover^[536] and retirement reliefs^[537] introduced on 1 July 1997. As Evans notes:^[538]

the greatest increase in personnel involved in CGT internal time was at the level of business proprietors and directors—those who would tend to have the greatest stake in determining whether the new reliefs were applicable and in accessing the reliefs where possible. It may be an ironic conclusion (but one that is nonetheless intuitively true) that the introduction of concessional treatment to the tax regime can often lead to increased compliance costs. The same can be true of choice—the more choice a taxpayer has, the greater the likely compliance costs as the taxpayer explores each possibility to obtain the optimal tax outcome.

However, the study found no significant increase in CGT compliance costs for personal (non-business) taxpayers.^[539] The number of personal taxpayers receiving capital gains increased from 270 531 in 1994–95 to 709 880 in 1997–98.^[540] More respondents obtained paid tax assistance as a result of CGT, up from 4.5 per cent to 5.2 per cent.^[541] Overall though CGT record keeping time fell from 2 per cent of total compliance time to less than half of 1 per cent.^[542]

Evans^[543] recently surveyed Australian tax practitioners regarding their attitudes to the compliance costs of the Australian CGT. Relevantly, he found that the survey showed that the compliance burden faced by personal taxpayers is directly influenced by the design features of CGT.

Relevantly, tax practitioners very strongly agreed that the ‘CGT legislation is complex’ (82 out of 94 respondents).^[544] Further, they also very strongly agreed that ‘compliance costs for CGT do not relate to the amount of the gain’ (84 out of 94 respondents).^[545] Importantly, they very strongly disagreed with the statement ‘compliance costs for CGT are lower now than they were five years ago’ (69 out of 94 respondents).^[546] Also, there was strong agreement for the statement ‘compliance costs for CGT do not closely relate to the size of the transaction’ (81 out of 93 respondents).^[547]

Further, Evans asked practitioners to assess the impact of 18 possible drivers for CGT complexity.^[548] The results were tabled as follows:^[549]

Rank	Factor	No.	Mean score	Frequency of citation (raw)	Frequency of citation (weighted)
1	(c) complexity of legislation	93	7.6	54	128
2	(q) the noof rules and exceptions	93	7.3	23	40
3	(g) Frequently changing legislation	93	6.9	28	49
4	(a) Record keeping requirements	93	6.6	27	64
5	(b) Poor legislative drafting	92	6.5	25	54
6	(i) Small business concessions	93	6.4	18	35
7	(d) Valuation issues	92	6.3	23	43
8	(n) Uncertainty of application	93	6.2	19	33
9	(m) Application to different entities	92	5.9	15	27
10	(k) Other reliefs and concessions	89	5.8	4	7
11	(r) The forms and instructions	90	5.8	10	22
12	(o) The no of calculations	93	5.6	4	6
13	(e) Calculating the cost base	93	5.5	8	14
14	(p) The difficulty of the calculations	91	5.1	2	3
15	(l) Identifying whether CGT applies	93	4.7	8	14
16	(j) The main residence exemption	93	4.5	1	1
17	(f) Calculating capital proceeds	93	4.4	1	1
18	(h) The CGT discount	93	4.0	1	1

The major driver of compliance costs was the complexity of legislation. Other important influences were the number of rules and exceptions, frequently changing legislation, record keeping requirements, poor legislative drafting and the small business concessions.

Evans concluded that the CGT compliance costs are high compared to the tax payable and the tax revenue collected.^[550] He also inferred from the tax practitioner responses that CGT compliance costs were higher in 2002 than five years ago, in 1997, prior to the enacted Ralph CGT reforms.^[551] Evans also concluded that CGT compliance costs are high in comparison to the costs involved with other taxes.^[552]

Thus these studies suggest that the CGT discount with its 28 pages of complex rules have added to the complexity and thus to the compliance costs of taxpayers. Although given the findings from Evans these extra costs may not be substantial.

On the other hand, the abolition of averaging would appear to simplify CGT since it removes an extra calculation. Similarly the freezing of indexation will remove time consuming indexation calculations for post 21 September 1999 assets. This appears to have reduced compliance costs.

The administration costs of a realisation CGT include the costs of tax design and drafting by Treasury, introducing and maintaining legislation by Parliament, and the administration start up costs of the Australian Taxation Office.^[553] The Australian Taxation Office will also incur costs for tax collection, assessing and auditing, debt recovery, forecasting tax revenue, advising, writing private and public rulings, dispute resolution and litigation.

There exists limited information about Australian administration costs given that the Australian Taxation Office does not provide a separate costing for CGT. Evans though provides the following table that shows the administration costs in the United Kingdom and Australia as a percentage of revenue yield.^[554]

Some caution must be exercised in relying upon these figures. Firstly, the UK figures were obtained from the annual reports of UK Inland Revenue and annual Taxation Statistics from the Australian Taxation Office.^[555] Thus other government administration costs^[556] incurred outside of the UK Inland Revenue and the Australian Taxation Office are excluded. Secondly, when interpreting UK figures it should be noted that such foreign research may not provide a good indicator for Australian tax administration given the differences in foreign tax regimes, taxpayer behaviour and tax administration. Thirdly, given the absence of a CGT costing by the Australian Taxation Office, Evans estimated the administration costs based on number of taxpayers with CGT compared to the overall number of taxpayers.^[557] Also, the CGT/yield ratio needs to be read carefully since it fails to take into account the integrity role of a CGT in preventing the transformation of ordinary income into exempt capital gains. As previously noted, a CGT prevents significant leakage from the income tax system.

The table firstly shows that for both countries the administrative costs of all taxes as a percentage of all tax revenue are falling over the eight-year period. Secondly, the table shows that CGT administration cost as a percentage of CGT revenue is highly volatile. This appears to be the result of the instability of CGT collections that move with the performance of the share and real estate markets. Thirdly, and most importantly, the administration costs of CGT as a percentage is significantly greater than the all taxes administration costs percentage. This is not surprising given the impact of CGT complexity on compliance costs.

Not surprisingly, the quantum of the impact of the CGT discount and other related reforms on administration costs is unknown. The CGT discount involved significant set up costs associated with tax design and drafting by Treasury, introducing and maintaining legislation by parliament and the administration start up costs of the Australian Taxation Office. Also, it can be inferred that the CGT discount and its complex rules would have contributed to ongoing administration costs.

The abolition of averaging would appear to simplify CGT administration since it removes an extra calculation. Also, the freezing of indexation has removed indexation calculations for post 21 September 1999 assets and this would also appear to reduce administration costs.

VIII CONCLUSION

On a positive note the reforms that abolished CGT averaging and that froze CGT indexation are to be welcomed. Both reforms worked to increase CGT revenue. Secondly, since both concessions only applied to a limited number of asset holders they did not assist neutrality and hence economic efficiency. Again, the limited availability of these concessions and the tax minimisation opportunities all damage horizontal equity. Further, vertical equity was also breached given the upside down effect of such an exemption and the concentration of wealth. Additionally, both provisions required cumbersome calculations and this added to CGT complexity and thus to operating costs. These concessions failed all four of the tax policy criteria.

However, the CGT discount has had a great and negative impact upon fiscal adequacy. Further, the CGT discount has breached horizontal and vertical equity since it is only available to asset holders. The loss of equity is serious given the great concentration of Australian wealth. Additionally, the discount adds to complexity given the numerous requirements and anti avoidance provisions totaling 22 pages of legislation. These all contribute to operating costs.

Whilst the Ralph Report asserted that this reform would enliven and invigorate the share market and achieve a better allocation of resources,^[558] other evidence suggests the economic impact is mixed. It may have reduced lock in but this does not appear to be a serious problem. The discount appears to have created inefficiencies through the tax advantages that provides to capital growth assets and thus damages neutrality.

Overall, this reform is not supported since it imposes great costs on fiscal adequacy and equity, as well as adding to complexity. Economic efficiency does not appear to be greatly assisted by this exemption.

^[1] Review of Business Taxation, A Tax System Redesigned, More Certain, Equitable and Durable, Report (July 1999) 104.

^[5] CGT averaging was abolished, Income Tax Rates Act 1986, Sch 7; and CGT indexation was frozen, Div 114 ITAA 1997.

The power to tax is the one great power upon which the whole national fabric is based. It is as necessary to the existence and prosperity of a nation as is the air he breathes to a natural man. It is not only the power to destroy, but the power to keep alive.

^[7] Australian Taxation Office, Australian Taxation Statistics 1999-2000 (2002) <www.ato.gov.au> 77.

^[8] Australian Taxation Office Annual Report 2000-2001 (2001) <www.ato.gov.au>, 120.

^[11] J Waincymer, Australian Income Tax principles and policy, (2^nd ed, 1993), 30.

^[12] R Krever and N Brooks, A Capital Gains Tax for New Zealand (1990) Victoria University Press, Wellington, 43, finds horizontal equity as a widely accepted and fundamental principle of social justice.

^[13] D F Bradford, Untangling the Income Tax (1986) Cambridge Mass, Harvard University Press 150–51.

^[15] J Norregaard, ‘The Progressivity of Personal Income Tax Systems’ in International Monetary Fund, Tax Policy Handbook , P Shome P, (ed) (1995) 130.

^[17] Ibid 112. Comprehensive income, defined as the sum, over an annual period, of a taxpayer’s current revenue less current costs, and the net change in value of the taxpayer’s assets and liabilities.

^[19] C Sandford, (ed) Further Key Issues in Tax Reform (1998) Fiscal Publications, Bath, UK, 204–07, explains the tax base broadening by governments in the 1980s as a reaction to the failure of the interventionist government policies of the sixties and seventies. A fundamental change in economic philosophy occurred, the overriding objective was economic efficiency and tax neutrality, leading the push to a broader income tax base.

^[20] Above n 1, 111. See also J Ralph, (Chairman of Committee) Review of Business Taxation, Third Report, A Platform for Consultation (1999) 28. The report points to the benefits of comprehensive income:

the imposition of income tax to raise revenue would have minimal impact on investment choices; the law could be far simpler and clearer; tax avoidance opportunities would be kept to a minimum. Business could focus on commercial choices rather than spend time and resources seeking to minimise liability.

While the comprehensive income tax base may never be implemented in practice, most successful market economies use that base in designing and evaluating the performance of their operating income tax systems ... Not surprisingly, the prospect of unbridled application of the conceptual comprehensive income tax base in all circumstances attracts taxpayer concern—especially in relation to liquidity, valuation, volatility, loss offset and effects on international competitiveness.

^[21] K W Asprey, (Chairman of Committee) Taxation Review Committee, Full Report, (1975) paras 3.23–3.26.

^[22] Australian Government, Reform of the Australian Taxation System, Draft White Paper Reform of the Australian Taxation system (1985) 15.

^[24] J Slemrod, ‘Professional Opinions About Tax Policy: 1994 and 1934’ (1995) 48 National Tax Journal 121: Slemrod reports on a tax policy opinion survey of 503 members of the National Tax Association with a response by 45 per cent academics, 32 per cent government employees, 28 per cent private sector. The survey repeats a 1934 senior American public finance professor’s survey verbatim (and asks additional questions).

^[26] United Kingdom Parliament Hansard, May 27 1853: col 722. In 1853 Gladstone said:

the Honourable Gentleman said that laws of this kind ought to be made intelligible to all persons who has not received a legal education. To bring the construction of these laws within the reach of such persons, was no doubt extremely desirable, but very far from being easy ... The nature of property in this country, and its very complicated forms, rendered it almost impossible to deal with it for the purpose of the income tax in a very simple manner.

H H Monroe, QC ‘Intolerable Inquisition? Reflections on the Law of Tax’ (1981) The Hamlyn Lecture Series 33. In 1980 the United Kingdom Special Commissioner of Taxation stated: ‘the plea today is that it would be some advance if laws of this kind were intelligible to those who have received a legal education’.

^[29] C Sandford, M Godwin, and P Hardwick, Administrative and Compliance Costs of Taxation (1989) 10.

^[30] B Tran-Nam, C Evans, M Walpole, and K Ritchie, ‘Tax Compliance Costs: Research Methodology and Empirical Evidence from Australia’ (2000) Vol LIII NTJ 229, 232–33.

^[33] Tran-Nam B., ‘Assessing the revenue and simplification impacts of the Governments tax reform’ (1999) JAT 329, 332-333.

^[34] Kenny P.L., ‘Capital gains taxation for New Zealand’ (2001) Vol. 7 No.4 New Zealand Journal of Taxation Law and Policy 265, 266.

^[35] Review of Business Taxation Report above n 1, 112, defined comprehensive income as ‘the sum, over an annual period, of a taxpayer’s current revenue less current costs, and the net change in value of the taxpayer’s assets and liabilities.’ Relevantly, all of the recent Australian income tax inquiries agreed, at least conceptually, on the neutrality advantages of a comprehensive income tax base. Unlike the current income tax system such a tax base would comprehensively tax capital gains and other forms of wealth. See Taxation Review Committee above n 21, 414; Australian Treasury Draft White Paper above n 22, 78.

^[42] Section115-25. Note s115-30 provides special rules about time of acquisition that deem earlier times of acquisition where assets are acquired under certain rollover provisions.

^[46] Subdiv 115-D. A listed investment company is an Australian resident listed company that has at least 90 per cent of its assets in allowed investments that include shares, units, options, rights, certain financial instruments and goodwill, s115-290.

^[58] Australian Taxation Office, Australian Taxation Statistics 1998-1999 (2001) <www.ato.gov.au>, 99; Australian Taxation Statistics 1999-2000 above n 7, 77; Australian Taxation Office, Australian Taxation Statistics 2000-2001 (2003) <www.ato.gov.au>, 120; Australian Taxation Office, Annual Report 2002-2003 (2003)

^[60] The Explanatory Memorandum to the New Business Tax System (Miscellaneous) Act (No 2) 2000 failed to estimate the revenue saving.

^[62] S Kelly, ‘Wealth on Retirement – Latest Estimates for Australia’, paper presented to the 9th Annual Colloquium of Superannuation Researchers, University of New South Wales, 9–10 July 2001, <www.natsem.canberra.edu.au>, 15.

^[65] R Woellner, S Barkoczy, S Murphy, and C Evans, Australian Taxation Law (2004) 1580–588, notes the tension that exists in defining tax avoidance, whether it involves tax planning, or whether it is confined to highly artificial arrangements. Rather than delineate these concepts in the context of CGT, this thesis adopts a broad definition of tax avoidance that includes tax planning, tax arbitrage (tax arbitrage occurs where taxpayers minimise tax by taking advantage of different tax consequences that apply to the various types of income, deductions, assets and types of entities) and highly artificial arrangements.

^[66] Ibid 1582. Defines tax evasion as ‘the non-payment of the tax which would properly be chargeable to a taxpayer if the taxpayer made a full and true disclosure of assessable income and allowable deductions’.

^[68] C Coleman, and L Freeman, ‘Cultural Foundations of Taxpayer Attitudes to Voluntary Compliance’, (1997) 13 Australian Tax Forum 311, 335–36; K A Kinsey, K W Smith, ‘Income Tax Cheating: Opportunities, Preferences and Sanctions’ American Bar Foundation. Working paper series #8718, presented at the 1986 meeting of the Midwestern Sociological Society, Des Moines, Iowa; C Sandford, ‘Why Tax Systems Differ: A Comparative Study of the Political Economy of Taxation’ (2000), 149.

^[70] Other CGT exemptions include: main residences, pre-20 September 1985 assets and death.

^[71] V Tanzi, The Individual Income Tax and Economic Growth an International Comparison, (1969) 83–84. He further notes that the effect of evasion means that all income and other taxes are higher.

^[73] F L Ayres, B R Jackson, and P A Hite, ‘The Economic Benefits of Regulation: Some Evidence from Professional Tax Preparers’ (1989) 64 The Accounting Review 300.

^[74] Y Grbich, ‘After Bellinz and Ralph, A New Focus for Decision Making in the Australian Tax System’ (2000) paper presented at the ATAX University of New South Wales 4^th International Conference on Tax Administration, Sydney, 28 –29 April 2000, 2.

^[76] Reform of the Australian Taxation System, above n 22, 78. See the following statement: ‘The lack of a general CGT represents a structural defect in the income tax system which lies at the core of many avoidance arrangements’.

^[77]Commonwealth, Senate Finance and Public Administration References Committee Inquiry in to the Business Taxation Reform, (1999), para 4.25.

^[78] L E Burman, The Labyrinth of Capital Gains Tax Policy, A Guide for the Perplexed (1999) 68.

^[79] Commonwealth, Parliament Hansard, House of Representatives, 11 September 1928, 6569, 6570

^[80] Section 16B of the Income Tax Assessment Act 1928 (Cth) was introduced to counter such tax avoidance schemes (now s 47 ITAA 1936).

^[84] Y Grbich, ‘Problems of Tax Avoidance in Australia’ in J G Head (ed) Taxation Issues of the 1980s (1983) 413.

^[86] The major asset class, residences are exempted by subdiv 118-B. Also, numerous other exemptions and rollovers apply for business assets, eg, see Divs 122, 124, 126 & 152. Also, death does not trigger CGT per Div 128. Further, numerous exemptions apply for non-residents per Div 136.

^[89] Commonwealth, Senate Finance and Public Administration References Committee Inquiry in to the Business Taxation Reform, (1999) para 4.31.

^[94] Wills v Petroulias (2003) NSWCA 286, para 12, ‘The charge before the magistrate ... was in the following terms: ‘‘That between about 1 September 1997 and 17 February 1999 Nikytas Nicholas Petroulias, also known as Nick Petroulias, did defraud the Commonwealth, namely the Australian Taxation Office contrary to s 29D of the Crimes Act 1914 in that, while an officer of the Australian Taxation Office he did, by dishonest means, assist taxpayers to avoid the payment of taxation”.’

^[95] C Sandford, Why Tax Systems Differ: A Comparative Study of the Political Economy of Taxation (2000) 149.

^[96] A Mitchell, ‘Wrecking Ball to Property Myths’ The Weekend Australian Financial Review (Sydney) 20–21 December 2003, 53. Mitchell noted that the Australian Bureau of Statistics found that almost 100 000 Australians were homeless on census night in 2001. Further, Mitchell stated that ‘[n]egative gearing and concessional taxation of capital gains have magnified the surge of investment in rental property’.

^[97] C Sandford, ‘Taxation and Social Policy: An Overview’ in C Sandford, C Pond, R Walker (eds), Taxation and Social Policy (1980) 4–5. Sandford found that tax concessions to house owners raise the value of homes providing capital gains to owners but reduce the capacity of would be purchasers.

^[98] F Anderson, and K Wheatley, ‘Rich May be Barred From First Home Buyer’s Grant’ The Advertiser (Adelaide) 19 December 2003, 6. Sourced from Productivity Commission estimates.

^[99] N Warren,‘No Easy Way to Unwind Negative Gearing’ (2003) Issue 2 Tax Matters, Atax, University of New South Wales, 7.

^[100] D Bassanese, ‘Too Many Property Investors’, Australian Financial Review (Sydney) 21–22 February 2004, 17–18.

See the statement: ‘There were nearly 2.0 million renter households in Australia in 1999, representing a quarter of all households’.

^[108] Australian Council of Social Services, ‘Tax Figures Show Growing Gap Between Rich and Poor’ ACOOS INFO 202, 21 January 2000, <www.acoss.org.au/info/2000/info202.htm>. This paper cites United Nations data which shows that Australia is one of the most unequal of all developed countries, having slipped from 7^th to 15^th place on their index of human development. There are two million people in Australia living below the poverty line; F Stillwell, Economic Inequality, Who Gets What in Australia (1993) 17. Stillwell notes research at the Social Policy Research Centre at the University of New South Wales, in conjunction with the international ‘Luxembourg study’, found that compared to eight other comparable nations, Australia ranks third highest in terms of the proportion of income in the top 20per cent of households. Also, Australia ranks third in terms of the proportion of income in the poorest 20 per cent of households. Consequently, Australia is among the most unequal of the countries studied.

^[109] P Saunders, ‘Examining Recent Changes in Income Distribution in Australia’ (2003) Social Policy research Centre discussion paper No 130, October 2003, http://www.sprc.unsw.edu.au/dp/DP130.pdf, 16. Saunders found that the degree of inequality has grown faster in the mid-1990s and speculated that market forces created the growing inequality as well as tax and social welfare rules.

^[115] A G A Faria, ‘Tax Coordination and Harmonisation’ in P Shome (ed) Tax Policy Handbook (1995), 231.

^[116] Evans Taxing capital gains above n 36, 116. Notes the exception of New Zealand.

^[117] Review of Business Taxation, An International Perspective, Information Paper, December 1998 (1998), 81–82.

^[120] Bradford above n 13, 148. Notes that ‘identifying an improvement in tax equity is difficult because there is no single measure of fairness ...’; G S Cooper, ‘The Benefit Theory of Taxation’ (1994) 11 Australian Tax Forum 397, 410. Cooper states: ‘Like all good ideas, the idea of equity is not a simple proposition but, rather, a complex idea capable of sustaining contradictory claims all alleged to be justified by it.'; H H Zee, ‘Taxation and Equity’ in P Shome (ed) Tax Policy Handbook (1995), 30–34.

^[121] See, for example, OECD, OECD Tax Database (2001) <http://www.oecd.org/dataoecd/43/63/1942474.xls> . This table shows that New Zealand’s top marginal personal income tax rate is 39 per cent, whilst in Denmark the top marginal personal income tax rate is 63.3 per cent (including social security taxes).

^[122] Zee above n 120, 30–34. Zee notes the many problems of measuring vertical equity.

^[124] A Reynolds, ‘Capital Gains Tax: Analysis of Reform Options for Australia’ (1999). A study commissioned by the Australian Stock Exchange for the Review of Business Taxation, <http://www.asx.com.au/shareholder/l3/MR160799_AS3.shtm> , 21.

^[125] Australian Taxation Statistics 1999–2000 above n 7, 77. There were 596 995 individual taxpayers with taxable incomes below $50 000 that received capital gains in 1999–2000.

^[128] Australian Taxation Statistics 1999–2000 above n 7, 77–78. Seven per cent of individual income taxpayers received capital gains in 1999–2000. Individuals with taxable incomes above $50 000 paid 67 per cent of all CGT paid by individuals.

^[129] Ibid 77–78. Also, note that only 7 per cent of individual income taxpayers received capital gains in 1999–2000.

^[131] Reform of the Australian Taxation System above n 22, 79–80. This paper noted this argument.

^[133] Report of Committee of Inquiry into Inflation and Taxation, Inflation and taxation (1975).

^[135] D W Jorgenson, and Y Kun-Young Tax Reform and the Cost of Capital (1991).

^[139] Reynolds above n 124, 8; M David, Alternative Approaches to Capital Gains Taxation (1968), 55–57. David expressed this argument by reference to rate of return formulas. He found though that the change in present value of an asset should be subject to CGT since the change generally responds to some value added to the original asset value by the owner or his or her agents. Even if the appreciation is a true windfall gain, there is no double taxation since upon sale the buyer obtains a real yield and amortises the cost base.

^[146] Reynolds above n 124, 9; virtually all countries restrict the deductibility of capital losses under realisation.

^[150] A J Auerbach, L E Burman, and J M Siegel, ‘Capital Gains Taxation and Tax Avoidance: New Evidence From Panel Data’, paper presented at a conference sponsored by University of Michigan office of tax policy research, October 24–25 1997 <www.urban.org/tax/capg11.html>, 129.

^[151] Section 102-5(1). Note, this restriction does not apply for CGT assets that do not qualify for the CGT discount, ie, indexed capital gains.

^[155] Sandford et al Taxation and social policy above n 97, 164; A J Auerbach, The Taxation of Capital Income 1983, 9. Auerbach states:

One point that is commonly made is that the introduction of a differential tax on capital income from different sources is deleterious to social welfare because it induces inefficiency in production arising out of a misallocation of capital.

^[156] Warren above n 99, 7, notes that gross rental income was $11.5 billion in 1999/2000. He notes the investment advantages of appreciating investment property, the immediate deductions for expenses including interest and the benefit of a 50 per cent CGT discount on any capital gain. This helps explain the prevalence of negative gearing and the low rental returns.

^[157] For example, an individual in the 2003–04 tax year has a top marginal income tax rate of 48.5 per cent (including Medicare levy), yet companies are only taxed at 30 per cent and superannuation funds are taxed at 15 per cent.

^[158] Interest and rent are included as assessable income per s 6-5 ITAA 1997. Dividends are included as assessable under s 44(1) ITAA 1936. Franking credits for dividends are included as assessable income under s 160AQT.

^[161] Ibid 43–44 provides a measure of the effective tax rate as the difference between the rates of return before and after all individual income tax expressed as a percentage of the before tax return.

^[165] Note that such a policy would be radical given that the income tax system is not indexed for inflation. Thus this issue is outside the scope of this paper.

^[166] CGT = net capital gain $61 x 48.5 per cent x 50 per cent CGT discount = $14.80

^[167] Burman above n 78, 47. 1 + real return = 1 + nominal return / 1 + inflation rate; (1.10/1.04= 1.0577)

^[168] Bonds annualised after tax return over 5 years is 5.1 per cent, thus the after tax real return is 1.1 per cent (1.051/1.04= 1.011)

^[169] Shares annualised after tax return over 5 years is 7.9 per cent, thus the after tax real return is 3.75 per cent (1.079/1.04= 1.0375)

^[171] Ibid. The effective tax rate = (real pretax return – real after tax return) / real pretax return; (0.10-0.011)/0.1 = 100 per cent.

^[173] As commentators have noted, an accruals CGT has liquidity and valuation problems that pose serious equity and complexity issues, and thus is not advocated in this paper. See, J Alworth, A Giampaolo, R Hamaui, ‘What’s Come to Perfection Perishes: Adjusting Capital Gains Taxation in Italy’ 2003 LVL 1(2) National Tax Journal 197.

^[177] P Groenewegen, ‘Taxation of Capital and Capital Gains: Reflections on Objectives and Potential Instruments’, paper presented at the CEDA and Australian Tax Research Foundation Seminar, 1 May 1985, 15.

^[180] R Krever, ‘The Continuing Debate Over Capital Gains Tax’ in S Ross, P Burgess, Income Tax: A Critical Analysis (2^nd ed, 1996), 67–68.

^[188] C Sandford, ‘Tax Reform of the Eighties in Retrospect: What Can We Learn?’ in C Sandford (ed), Further Key Issues in Tax Reform (1998) 206.

^[193] M K McGee, ‘Capital Gains and New Firm Investment’ (1998) LI(3) National Tax Journal 653, 654.

^[195] M McIntyre, F E A Sander, D Westall editorial comment in M McIntyre, F E A Sander, D Westall (eds) Readings in Federal Taxation, (2^nd ed, 1983) 400.

^[197] R H Gordon, J Slemrod, ‘Do We Collect any Revenue From Taxing Capital Income?’ in J Poterba (ed) Tax Policy and the Economy Vol 2 (1988), 89–130.

^[205] South African Department of Finance ‘Guide to Capital Gains Tax’ <www.sars.gov.za>, 4.

^[209] A J Auerbach, ‘Measuring the impact of tax reform’ (1996) XLIZ National Tax Journal 665, 671–72.

^[213] E M Uslaner, ‘Democracy and Social Capital’ in M E Warren (ed) Democracy and Trust (1999), 122. Social capital refers to ‘features of social organization, such as core values and norms (including social trust) and networks, that facilitate coordination and cooperation for mutual benefit.’

^[214] J Slemrod, ‘On Voluntary Compliance, Voluntary Taxes and Social Capital’ (1998) LI National Tax Journal 485, 485.

^[218] R Hjerppe ‘Social Capital and Economic Growth’ from the Government for Institute of Economic Research Finland, presented at the International Conference on Social Capital, Tokyo 24–25 March 2003, <http://www.esri.go.jp/en/workshop/030325/030325paper2-e.pdf> , 20.

^[219] Australian Productivity Commission, Research Paper ‘Microeconomic reforms and Australian Productivity: Exploring the Links’ 12 November 1999, Volume 1, <http//www.pc.gov.au>, 160. The Australian Productivity Commission reviewed international studies into national wealth, finding that the most important determinants of wealth were policy, openness to trade, rule of law, quality of institutions, geography, resource endowments and human capital. Interestingly taxation and the level of savings and investment were not directly highlighted.

^[225] A voluntary tax refers to a tax that has no withholding taxes. The taxpayer voluntarily declares the tax liability to the tax administrator. Under the self-assessment system of income tax, CGT is voluntary since it requires a taxpayer to record all net capital gains in their annual income tax return.

^[229] Unattributed, ‘Taxpayer Survey’ Adelaide, The Advertiser (Adelaide) 9 January 2001, 7.

^[232] J Stotsku, ‘Summary of IMF Tax Policy Advice’ in P Shome (ed) Tax Policy Handbook (1995) 281.

^[235] OECD, ‘OECD Tax Database’ (2001) <http://www.oecd.org/dataoecd/43/63/1942474.xls> . On average the top marginal income tax rate for labour income for 30 OECD countries was 50.1 per cent in 2001. This rate included marginal personal income tax and social security contribution rates.

^[237] Ibid 54. Other factors were: technological advances, physical capital, firm organization, management practices and work arrangements, resource allocation, scale scope and specialisation, and plant – firm turnover within industries.

^[238] D J Boyer, S M Russell ‘Is it Time for a Consumption Tax?’ National Tax Association Symposium, A new Agenda–Old Constraints, Arlington, VA, 363, 364;

^[243] Ibid; OECD,‘OECD in Figures: Statistics on Member Countries’ (2000) <www.oecd.org>, 87. Australia’s spending in investment in knowledge, that is public spending on education, research and development expenditure and software, of 6.7 per cent of GDP ranks below the OECD average of 8 per cent of GDP, implying a need for a greater focus on human capital.

^[251] W C Randolph, D L Rogers, ‘The Implications for Tax Policy of Uncertainty about Labor-Supply and Savings Responses’ (1995) 48 National Tax Journal 429, 431.

^[259] J G Gravelle, 'Behavioural Feedback Effects and the Revenue Estimating Process' (1995) National Tax Association Symposium, A New Agenda–Old Constraints, Darlington, VA, 463, 468.

^[264] F Sammartino, D Weiner, ‘Recent Evidence on Taxpayers’ Response to the Rate Increases in the 1990s’ (1997) L National Tax Journal 683, 683–84.

^[269]American Council for Capital Formation, ‘Overcoming Barriers to US Economic Growth, International Competitiveness, and Retirement Security’ (1999) <www.accf.org/June99test.htm>, 16, M Feldstein (ed), Capital Taxation (1983) 1.

^[272] Commonwealth, House of Representatives, Budget Strategy and Outlook 1997-98 Budget Paper No 1 1997 (P Costello, Treasurer; J Fahey Finance Minister), 1–12.

^[273] Commonwealth, House of Representatives, Budget Speech 1997 (P Costello, Treasurer) 2.

^[275] W G Gale, J Sabelhaus,‘The Savings Crisis: In the Eye of the Beholder?’ (1999) The Milken Institute Review, 26 August 1999,

<http://www.brook.edu/views/articles/gale/19990826.htm> , 2. Examines a number of ways of defining savings, before providing a simple definition of savings equals income minus consumption. Although they note that this definition flounders with the defining of what is ‘income’.

^[278] Notably the CGT discount for superannuation funds is only 1/3^rd compared to the 50 per centindividual discount.

^[280] M Lally, ‘The Cost of Capital Under Dividend Imputation’ (2002) prepared for the Australian Competition and Consumer Commission <http://www.accc.gov.au/gas/br_reg_iss/Cost_capital_dividend.pdf> , 9. Lally notes that in 1996, foreign investors hold around 30 per cent of shares in the Australian equity market.

^[281] Lock in refers to the impact of a realisation CGT on investors that encourages them to hold onto assets in order to defer tax.

^[288] Burman above n 78, 56, states ‘available evidence suggests that, in the aggregate, saving does not respond to the rate of return’.

^[296] R M Gordano, a Wall Street economist, commented on E Steuerle’s paper ‘Effects on Financial Decision Making’ in J A Pechman (ed) Tax Reform and the United States Economy (1986) 67.

^[308] M Feldstein, and S Yitzhaki, ‘The Effects of the Capital Gains Tax on the Selling and Switching of Common Stock’ in M Feldstein (ed), Capital Taxation (1983) 237, 250. The study indicated that reducing the tax on capital gains on corporate stock would increase realisations and tax revenue.

^[350] M B Schmidt, ‘Savings and Investment in Australia’ (2003) 35 (1) Applied Economics 99, 99–100.

^[351] D P Chaudhri, E Wilson, ‘Savings, Investment, Productivity and Economic Growth of Australia 1861–1990: Some Explorations’ (2000) 76 (232) Economic Record 55, 55–57.

^[358] McGee above n 193, 653. McGee found that a CGT rate cut is not a well targetted means of encouraging new firm capital formation.

^[366] Gravelle ‘The economic effects of taxing capital income’ above n 259, 15.

^[369] American Council for Capital Formation, ‘Capital Gains Taxes and United States economic growth; A Retrospective Look’ (1999) <www.accf.org/capgainsDR1799.htm>.1-2. Burman above n 78, 53: the cost of capital is the before rate of return a business must earn on new investments.

^[370] American Council for Capital Formation Capital Gains Taxes above n 370, 1–2.

^[372] A Goolsbee, 'The Importance of Measurement Error in the Cost of Capital' (2000) LIII National Tax Journal 215, 215.

^[377] This can be inferred from the fact that superannuation funds provided 38 per cent of reported net capital gains in 1999–2000, and 45 per cent in 2000–2001, Australian Taxation Statistics 2000–2001 above n 58, 123–24.

^[382] C R Hulten ‘Comments on a Paper by L L Dildine ‘Effects on Industry’ in J A Pechman (ed) Tax Reform and the United States Economy (1986) 50.

^[387] Gann above n 270, 93; she also notes though that increased capital formation will only have a modest impact on growth; and at 81 finds a substantial consensus that income tax needs to be modified to achieve increased neutrality in the taxation of capital.

^[388] American Council for Capital Formation Overcoming barriers above n 269, 6.

^[404] S M Fazzari, 'Capital Income Taxes and Economic Performance' (1998) 91^st National Tax Association Conference Proceedings, Austin, Texas 433, 434.

^[418] OECD,‘OECD in Figures: Statistics on Member Countries’ (2000) <www.oecd.org>, 14, 15, 78; Ernst and Young, ‘Doing Business in’ (2000) <www.doingbusinessin.com>, Czech Republic, Italy, New Zealand, Netherlands and Greece have minimal taxation of capital gains.

^[426] Section 136-25, shares in Australian public companies are generally exempt from CGT for foreign investors; also see Lally above n 280, 9. Lally notes that in 1996, foreign investors hold around 30 per cent of shares in the Australian equity market. He finds that there is virtually a free flow of equity capital between Australia and the world’s principal equity markets.

^[427]Australian Stock Exchange, ‘Australian market overview’ above n 426. This report relevantly states ‘Australia has deep and efficient markets in equity securities and related derivative products. In terms of market capitalisation, Australia's premier stock exchange, the ASX, is currently the twelfth largest stock market in the world and the third largest in the Asia-Pacific behind Japan and Hong Kong’.

^[433] Kelly above n 62, 13–15. This research also shows a concentration of Australian wealth, with 10 per cent of individuals holding 43 per cent of Australian wealth. They hold the wealth in diversified assets such as: interest bearing deposits (13 per cent); shares and other investments (14 per cent); home (30 per cent); rental properties (13 per cent); business (18 per cent); and superannuation (13 per cent).

^[437] D Futrelle ‘A Penny for Your dotcom’ (2003) 32(9) Money 28 (New York) 28.

^[448] D Chua, J R King, ‘The Mechanics of Integration’ in P Shome (ed) Tax Policy Handbook (1995) 151.

^[451] Ibid, notes that such a system would require a vast amount of information about geographically diverse spread of owners, difficulties in attributing retained earnings, problems in tracing of ultimate owners, and has liquidity issues.

^[458] Slemrod ‘Professional opinions about tax policy’ above n 24, 124-132.

^[460] American Council for Capital Formation ‘Overcoming barriers’ above n 269, 4.

^[462] American Council for Capital Formation ‘Overcoming barriers’ above n 269, 4.

^[466] P A Gompers, J Lerner, ‘What Drives Venture Capital Raising?’ in M N Baily et al (eds) Brookings Papers on Economic Activity : Microeconomics (1998) 149.

Superannuation funds have $218 billion invested in Australian equities and units in trusts as at June 2002; Australian Stock Exchange above n 426. At the end of June 2002, the market capitalisation was $700 billion.

^[477] Lally above n 280, 9. Lally notes that in 1996, foreign investors held around 30 per cent of shares in the Australian equity market.

^[481] Head Capital gains taxation above n 175, 150–51; Groenewegen above n 177, 15,

^[484] M L Louie, ‘Realizing Appreciation Without Sale: Accrual Taxation of Capital Gains on Marketable Securities’ (1982) 34 Stanford Law Review 857, 867.

^[485] Kelly above n 62, 12, The Australian wealth held in shares and other investments increased in 1998 dollars from $40 billion in 1986 to $120 billion in 1998. The Australian wealth held in net businesses increased in 1998 dollars from $103 billion in 1986 to $222 billion in 1998.

^[486] Burman above n 78, 72, citing A Sandmo, ‘The Effects of Taxation on Savings and Risk Taking’, in J Alana, A J Auerbach, M Feldstein (eds), Handbook of Public Economics, vol 1 (1985) 307.

^[488] Reynolds above n 124, 5; B Howe, B Howell, ‘Understanding the New Venture Capital Concessions’ (2003) 7(1) Tax Specialist 3, 3. Howe and Howell note that Australia was traditionally considered to be a high tax jurisdiction for foreign venture capitalists.

^[489] Evans ‘Taxing capital gains’ above n 36, 133. Evans modelled the CGT impact for an unmarried individual on twice average earnings with a capital gain of US$40 000. In Australia the effective tax rate was 47 per cent (ignoring Medicare levy), in Canada 38 per cent, in UK and US both 20 per cent , in Ireland 19 per cent and in New Zealand 0 per cent. The average tax payable in all countries was US$8068 whilst the CGT amounted to $15 801 in Australia.

^[490] J A Pechman in J A Pechman (ed) World Tax Reform: A Progress Report (1987), 253.

^[492] B McLean, S Shrestha, ‘International Financial Liberalisation and Economic Growth’ Reserve Bank of Australia Research Discussion Paper (2002), <http://www.rba.gov.au/PublicationsAndResearch/RDP/RDP2002-03.html> . This paper concluded that both foreign direct investment and portfolio inflows enhance economic growth.

^[497] Section 136-25 sets out the assets that have the necessary connection with Australia that non-resident investors may be liable to CGT. These include: land and buildings located in Australia; a CGT asset used in carrying on a permanent establishment; a share in a private company or trust that is a resident; a share in a public company that is a resident in which the taxpayer owns more than 10 by value of the shares; a share in a unit trust that is a resident in which the taxpayer owns more than 10 per cent by value of the units.

^[502] J M Weiner, H J Ault, ‘The OECD’s Report on Harmful Tax Competition’ (1998) National Tax Association Symposium: What do We Mean by Taxpayer Relief, Arlington, VA, 601, 601.

^[503] J Pope, ‘The Compliance Costs of Major Taxes in Australia’ in C Sandford (ed) Taxation Compliance Costs Measurement and Policy (1995) 102.

^[504] C Evans ‘The Operating Costs of Taxing Capital Gains – A Conspectus’ in C Evans, J Pope, J Hasseldine (eds) Tax Compliance Costs: A Festschrift for Cedric Sandford (2001) 164.

^[505] I Wallschutzky, B Gibson, Small Business Cost of Compliance Project: Final Report (1993) Australian Taxation Office, 29.

^[509] Evans et al, A report into the Incremental Costs of Taxpayer Compliance (1996) Australian Government Publishing Service. Evans et al were commissioned by the Australian Taxation Office to examine the incremental costs of compliance so as to calculate the impact of law reform on compliance costs. This 1994–95 year study relied on three extensive mail surveys of 10 000 taxpayers. The study though made no calculation of compliance costs but sought to identify methodologies and information needed to make such a computation.

^[510] C Evans, K Ritchie, B Tran-Nam, M Walpole, A Report into Taxpayer Costs of Compliance, (1997) Australian Government Publishing Service.

^[511] Evans ‘The operating costs of taxing capital gains’ above n 505, 165.

^[526] Evans et al ‘A report into taxpayer costs of compliance’ above n 511, 54–57.

^[527] Ibid at 17, 31, Compliance costs equated to the costs of taxpayers complying with tax laws, including external advisers and the taxpayer’s own time and costs, less tax deductibility benefits, and less cash flow benefits. This compliance costs equation though does not reflect the following compliance benefits. Firstly, compliance with CGT will provide managerial benefits through improved record keeping. Secondly, these compliance costs are offset by temporary and permanent tax savings provided by the deferral benefit (that is the benefit of deferring the payment of CGT at the time of realisation) and CGT exemptions (for example the small business CGT reliefs in Division 152).

^[528] Evans ‘The operating costs of taxing capital gains’ above n 505, 167.

^[529] Ibid 167–68.

^[530] Ibid.

^[531] Ibid.

^[532] Ibid.

^[533] Ibid.

^[534] Ibid.

^[535] Ibid.

^[536] Ibid.

^[537] Ibid.

^[538] Ibid 168.

^[539] Ibid.

^[540] Ibid.

^[541] Ibid.

^[542] Ibid.

^[543] C Evans ‘Taxing Personal Capital Gains in Australia: Causes of Complexity and Proposals for Reform’, paper presented at the Australasian Tax Teachers Association National Conference, Flinders University, Adelaide, 29–30 January 2004, 1. Over 300 practitioners were surveyed.

^[544] Ibid 19–20.

^[545] Ibid.

^[546] Ibid 20–21.

^[547] Ibid.

^[548] Ibid 21–22.

^[549] Ibid 22.

^[550] Ibid 24.

^[551] Ibid 24.

^[552] Ibid 25.

^[553] Tran-Nam et al ‘Tax compliance costs’ above n 30, 332–33.

^[554] Evans ‘The operating costs of taxing capital gains’ above n 505, 171.

^[555] Ibid.

^[556] For example, Parliament and Treasury costs.

^[557] Evans ‘The operating costs of taxing capital gains’ above n 505, 171.

^[558] Review of Business Taxation Report above n 1, 598.

AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/JlATaxTA/2005/11.html

Year	Net capital gains $ m	Capital Gains Tax revenue $ m
1998–99	17 490	4356
1999–2000	19 803	5330
2000–01	16 924	4423
2001–02	12 465	3657

Year	Estimated tax expenditure from the CGT discount ($ billion)
2000–01	1.31
2001–02	1.81
2002–03	1.81
2003–04	1.81
2004–05	1.91
2005–06	2.03

City	Value June 1998	Value June 2003	Change %
Sydney	249 000	388 000	56
Melbourne	155 000	316 000	104
Brisbane	144 000	277 000	92
Perth	143 000	206 000	44
Adelaide	119 000	212 000	79
Hobart	107 000	173 000	61
Darwin	174 000	211 000	21
Canberra	156 000	285 000	83
Average	155 875	258 500	66

Type of capital income	Assessable income $ billion	Percentage
Gross dividends	7.6	40
Realised net capital gains	5.9	31
Gross interest	5.4	28
Net rent	0.2	1
Total	30.4	100

Year	Bond $	Yield $	48.5% Tax $	Shares $	Yield $	48.5% Tax $
0	100			100
1	105.2	10	4.9	110	10.0
2	110.6	10.5	5.1	121.0	11.0
3	116.3	11.1	5.4	133.1	12.1
4	122.2	11.6	5.6	146.4	13.3
5	128.3	12.2	5.9	161.1	14.6	14.8
Net value After tax	128.5			146.2

Journal of the Australasian Tax Teachers Association